Skip to main content

Full text of "The new conquest of central Asia : a narrative of the explorations of the Central Asiatic expeditions in Mongolia and China, 1921-1930"

See other formats













: 3 


4Py FOR THE ^^. 


■< FOR ^ 


P^ FOR /^ 



From iJ pinntiny by Arthur A. Jansson 











WALTER GRANGER, Sc.D., Chief Pal/eontologist 

CLIFFORD H. POPE, B.S., Herpetologist 

NELS C. NELSON, M.L., Arch/eologist 





With 128 Plates and 12 Illustrations in the 
Text and Three Maps at end 



CHESTER A. REEDS, Ph.D.. Editor 





Copyright, 1932 


The American Museum of Natural History 

Published, December 29, 1932 

First Edition 

All rights reserved. This book, or parts thereof, must 

not be reproduced in any form without permission. 

Made in the United States of America. 




Through Their Interest in Science and 

Exploration Made Possible the 

Central Asiatic Expeditions 

This Booli 


Gratefully Dedicated 


This book is an introduction to the scientific publications on the explora- 
tions of the Central Asiatic Expeditions in Mongolia and China, from 1921 
to 1930. It tells how the work was done and attempts to give a picture of 
the country itself, of the Expeditions' daily life in camp and on the trail, of 
the transport by camels and motor cars, and of the principal discoveries in the 
various branches of science represented. It has been written to give the layman 
a comprehensive view of the Expeditions and makes no attempt to discuss the 
discoveries in detail, inasmuch as that is done in the series of scientific volumes 
which follow. 

For much technical information I have drawn freely upon the Preliminary 
Reports of the Expeditions' work, published in the American Museum No- 
vitates; also upon Volume 11, "Geology of Mongolia," by Professors Berkey 
and Morris, and upon Volume IV, "The Permian of Mongolia," by Professor 
Grabau, in this series. To the authors of all these publications I owe my best 

The permanency and value of any exploration lie in its published reports. 
Long after the field work has been forgotten, these volimies will remain as a 
record of accomplishment. 

The success of the work was due to no one man. The Expedition of each 
of the various years was a well-balanced machine in which every member was 
a vital part. Had any individixal failed in his allotted task, the efficiency of 
the machine would have been impaired, if not fatally crippled. The services 
of brilliant scientists having been obtained, the way in which they cooperated 
as a imit and worked for one ideal without friction insured success. During 
all the years we were in the field we lived as a happy family without a single 
quarrel. This is an almost unique record and one of which we are justly 
proud. No finer, more loyal or more devoted group of men could ever be 
brought together. The privilege that I had of organizing the Expeditions 
and leading them in the field and the personal loyalty of the men to me will 
ever remain as my most treasured memories. 

To those individuals and institutions who financed the Expeditions, we 
owe a debt which we could attempt to pay only by utilizing to the utmost the 
opportunities which they so generously afforded us. Particularly are we 


indebted to Asia magazine, New York, and to the Field Museum of Natural 
History, Chicago. They have our deepest gratitude. 

Without the prophetic vision and the enthusiastic interest and assistance 
of Professor Henry Fairfield Osbom, President of the American Museum of 
Natural History, the Expeditions never could have become an accomplished 
fact or continued to operate. The Board of Trustees of the American Mu- 
seum, individually and collectively, gave us the most loyal support in every 
way. To Director George H. Sherwood we feel a peculiar gratitude; upon his 
shoulders fell much of the administrative work of the Expeditions while we 
were in the field; to him we turned for every conceivable detail, personal as 
well as official, and always received a cheerful response. Mr. Frederick H. 
Smyth, Bursar of the Museum, handled the financial matters in his usual 
efficient manner. The former American Minister to China, Mr. J. V. A. 
MacMurray, was my guide and confidant in carrying on the diplomatic 
negotiations with the various Chinese governments which year by year became 
increasingly difficult. He gave imstintingly of his time and thought to our 

To Doctor Walter Granger, I owe a great personal debt. As Second in^ 
Command of the Expeditions, we worked together in the most complete 
harmony. Without his unselfish support and his profound knowledge of 
palaeontology and field work, the Expeditions never could have achieved the 
degree of success which rewarded our explorations. Moreover, he has given 
freely of his time and thought in the preparation of this volume. 

I am also indebted to Doctor Chester A. Reeds for his arduous editorial 
work which has greatly enhanced the value of the book. 

It is impossible to mention by name the hundreds of individuals and 
business organizations that rendered us substantial aid. IMany of them are 
referred to in the body of the work. Without their generous response and 
assistance our task would have been made much more difficult. 

It is a source of the deepest regret that, after the Nationalist Party 
obtained control of China in 1928 and the capital was changed from Peking 
to Nanking, the cordial relations, which had existed under the Peking govern- 
ment between the Expeditions and Chinese institutions and officials, became 
increasingly strained. With the advent of the new government, all projects 
concerning foreign explorations were referred to the Society for the Preser- 
vation of Cultural Objects. 

In September, 1928, this society, later known as the National Commis- 
sion for the Preservation of Ancient Objects, seized our collections upon 
their arrival in Kalgan. A period of negotiations followed between 
Washington, Nanking, Peking and the Commission. After six weeks of effort 
our collections at Kalgan were finally released. 


Following the seizure of the Expedition's collections, the Commission for 
the Preservation of Ancient Objects continued to oppose the work not only of 
our Expedition, but also of other foreign expeditions. As a vesvlt of this oppo- 
sition our work was suspended in the year 1929, restricted in 1930, and brought 
to a full stop in 1931, despite the continued efforts of President Osbom of the 
American Museum of Natural History in New York, Secretary of State 
Stimson at Washington and representatives of our State Department in 
China. Notwithstanding this present opposition on the part of Chinese 
institutions and officials, it is to be hoped that the dawn of a more tolerant 
era of sympathy and cooperation with foreign scientific endeavor will not 
be long delayed. 

Roy Chapman Andrews. 

The American Museum of Natural History, 
New York, N. Y. 
April 29, 1932. 




Andrews, Roy Chapman. — Leader and Zoologist, 1921-1930. 

Badmajapoff, T. — Mongolian Political Representative, 1922. 

Beckwith, Radcliffe H. — Geologist, 1926-1927. 

Berkey, Charles P. — Chief Geologist, 1922-1925. 

Butler, F. B. — Assistant Topographer, 1925. 

Chaney, Ralph W. — Palasobotanist, 1925. 

Colgate, S. Bayard. — Chief, Motor Transport, 1922. 

Eriksson, Joel. — Agent in Mongolia, 1926-1930. 

Garber, A. Z. — Siargeon, 1930. 

Grabau, A. W. — Research Associate, 1922-1930. 

Granger, Walter. — Chief Palasontologist, Second in Command, 192 1 -1930. 

Hill, W. P. T.— Topographer, 1928. 

Horvath, G. — Motor Transport, 1928. 

Johnson, Albert F. — Assistant in Palaeontology, 1923. 

Johnson, C. Vance. — Motor Transport, 1923. 

Kaisen, Peter C. — Assistant in Palaeontology, 1923. 

Larsen, F. A. — Interpreter and Expedition Agent, 1922-1926. 

Loucks, H. A. — Surgeon, 1925. 

Lovell, Norman. — Motor Transport, 1925. 

Matthew, W. D. — Palaeontologist, 1926. 

Morris, Frederick K. — Geologist, 1922, 1923 and 1925. 

Nelson, Nels C. — Archaeologist, 1925. 

Olsen, George. — ^Assistant in Palaeontology, 1923-1925. 

Perez, J. A. — Surgeon, 1928. 

Pond, Alonzo W. — Archaeologist, 192S. 

Pope, Clifford H. — Herpetologist, Chinese Division, 1921-1926. 

Roberts, L. B. — Chief Topographer, 1925. 

Robinson, H. O. — ^Assistant Topographer, 1925. 

Shackelford. J. B. — Photographer, 1922, 1925, 1928. 

Spock, L. Erskine. — Geologist, 1928. 

Teilhard de Chardin, P^re. — Geologist, 1930. 

Thomson, Albert. — Assistant in Palaeontology, 1928-1930. 

Wyman, W. G. — Topographer, 1930. 

Young, J. McKenzie. — Chief, Motor Transport, 1923-1930. 


Preface ...... 

Scientific and Technical Stafi 

List of Figures ..... 

List of Plates ..... 




Conception of the Central Asiatic Expeditions 

The growth of the idea 

Financial support 

Why Mongolia? 

Physical difficulties to be considered 

Motor transport as a solution . 
Political difficulties 
The method of correlated work 

Intensive scientific investigation 

Reconnaissance and specialization 

Field units 

Laboratory arrangements 
The staff 
Motor transport 


Gasoline cases 
Camp equipment 


Clothing . 















Vindication of plans by field results 
Weather conditions 
Mortality among camels 
Contacts and cooperation 

Headquarters at Peking 
Collecting in China . 
Final preparations at Peking 
Kalgan, the advance base 
The main party at Kalgan 

The start from Kalgan 
Geology of the Wan Ch'uan Pass 
Through the Great Wall and beyond 
Chinese farm lands of Inner Mongolia 
Birds of the southern grasslands 
Geology of the southern grasslands 
Previous palccontologic work near Hallong 
The telegraph road 
Expedition activities . 
Mammals of the southern grasslands 


The Gobi erosion plane 

In the vicinity of P'ang Kiang 

The Gobi life zone 

Southern Mongol tribes 

P'ang Kiang to Iren Dabasu 

The first evidence of fossils 

The Iren Dabasu basin 

Later Sediments of the Iren Dabasu basin 

Iren Dabasu to Ude . 

Jurassic strata near Ude 

Ude to Sain Usu 

Well-water in Mongolia 





Mammals near Sain Usu ....... 44 

With the caravan at Tuerin . 

• 44 

Caravan customs 

. 46 

False spring weather . 

. 46 



Faunal zones between Kalgan and Urga 

• 47 

Fauna of the Tuerin area 


Wolves and foxes 

. 48 

Birds .... 

• 48 

Tuerin monastery 


Lamaism and its influence in Mongolia 


Mongol warfare 


From Tuerin toward Urga . 


Northern grasslands . 


Bogdo Ola forests 


Tola River .... 




The Sacred City of the Living Buddha 


Russianization of Urga 


Diplomatic negotiations 


Festival of the Maidari 


Dress of the northern Mongols 


Relics of the Urga that has disappeared 


Agreement with the Urga government 


The "Allergorhai horhai" 


A good place to leave 




Leaving Urga . . . . 


Rendezvous at Bolkuk Gol . 


Tola Valley .... 


Eye-strain in the desert 


"Nigger heads" 


Five Antelope Camp . 


Faima at Five Antelope Camp 





ArgTil and camp cookery 

Southwestward toward Tsetsenwang 

Snakes .... 

Monastery of Tsetsenwang . 

The Great Mongolian Bathylith 

Fauna near Tsetsenwang 

Prince Tse Tsen 

Cheese and kumiss 

Traces of pre-Mongol people 

Jurassic plant fossils . 

Cara\'an plans 

Southwestward from Tsetsenwang . 
Canyon Brook .... 

Faxma at Canyon Brook 
Caravan trails . . . 

Desert weathering of granite 
The Ongin Gol .... 

The story of a massacre 
Rainy Gulch Camp . . 

Sain Noin Monastery 
Geology of the Sain Noin area 


Forest Camp ..... 

Mongol children .... 

The forest of northern Mongolia 

Toward the desert from Forest Camp 

The Hot Springs of Arishan . 

The Prince of Sain Noin 

The valley of the Sacred Mountain 


Southward from the Hot Springs of Arishan 
The basin of Gorida .... 



The "Post Road" 

The easternmost Alta ranges 

Across country toward Mount Uskuk 

A lake of salt .... 

Badlands near Mount Uskuk 

Camp Ondai Sair 

Fossils from the Cretaceous Ondai Sair formation 

Fauna at Camp Ondai Sair . 

Geological map of the Moiant Usloik region 

Wild Ass Camp .... 

A Baluchitherium skull 

A fossil spadefoot toad 

Fossil-hunting in the badlands of Loh 

Simimary of the Hsanda Gol fauna 

The geologic relation of the beds in the Tsagan Nor basin 

Securing the first specimen 
Speed and endurance 
Food and water 
Colts . 
Range and habits 

Typical desert faima at Loh 
Black vtiltures 
Eagles and kites 
A Fovuth of July in the desert 
A trip to Urga 
On the shore of Tsagan Nor 
Favma at Tsagan Nor 
Desert gazelles 

A Mongol encampment at Tsagan Nor 
Inside a Mongol yurt .... 

Mongol food ...... 















Mongol flocks and herds 

The Mongol on horseback . 

Sports of the Mongols 

An annual field meet .... 

Mongol characteristics 

Mongol hospitality .... 

Mongol dogs ..... 

How news travels in Mongolia 

The Mongols' sense of direction 

Mongol and Chinese attitudes contrasted . 

Sheep-shearing and felt-making 

Mongol women and children 

Mongol manners .... 

The pastoral life of the Mongols 

Lamaism and the decline of the Mongol race 

Recent political events 

"Base camp at Tsagan Nor . 
"Bones as large as a man's body" 
Tamarisk and dune berries . 

Completion of the geological map 
Bones of a giant Baluchitherium 
Pliocene fauna of the Hung Kureh formation 
Homeward from Tsagan Nor 
Surface water and waterfowl 
A desert oasis . 
Camp Volcanic CliflFs 
Mongols at Artsa Bogdo 


The Artsa Bogdo range 
A hunting trip for sheep and ibex 
Prehistoric pictures . 
Caravan afifairs 







13 s 







Ibex and bighorn sheep 

Artsa Bogdo faima .... 

Geological reconnaissance south of Artsa Bogdo 

The first Palaeozoic fossils 

Geologic structure of Artsa Bogdo . 

Oshih Basin ..... 


Eastward from Artsa Bogdo 
Fossils at Shabarakh Usu 
The Flaming Cliffs . 
Northward from Shabarakh Usu 
The "Hundred-Mile Tennis Court" 
Ongin Gol-in-Sumu 
The Sair Usu Trail . 
Sair Usu 

First Palffiozoic strata in place 
Mongols' fear of strangers 
Difficulties of the trail 
Ardyn Obo 

The Ardjm Obo formation . 
Prospecting for fossils at Ardyn Obo 
Fossil fauna of the Ardyn Obo formation 
An eagle's nest 
The approach of winter 


The departure from Ardjm Obo 

The Jisu Honguer formation 

The Shara Murun fossil beds 

Ula Usu and Baron Sog-in-Sumu 

The Shara Mimm River 

A snowstorm . 

Chinese frontier in the southern grasslands 

The return to Kalgan 










Summary of the 1922 season's work 

In perspective 

The traverse 

The geology 

The palaeontology 

The zoology 

The photography 

The motor transport 
Plans for 1923 
Safe return of the caravan 
Fossil pits of Yen-ching-kou 

Staff of the 1923 Expedition . 
Back to Mongolia .... 
Scenes along the Kalgan-Urga road 
The whereabouts of our caravan 
Food and characteristics of a Mongolian camel 
Fossil collecting at Iren Dabasu 
The Iren Dabasu "quarries" and their fauna 
The Houldjin gravels and their faima 


An experience with brigands 

The return to Iren Dabasu . 
Irdin Manha in 1923 . 
A terrific windstorm . 
The subdivisions of the Irdin Manha beds 
The fossil fauna of the Irdin Manha beds 
The livnng faima at Irdin Manha . 
The 1923 passports for Outer Mongolia 
On the way to Ula Usu 
The camp at Ula Usu 
The Shara Munm beds 
A pariah dog . 
A full gale in the desert 
The result of ten months of drought 









The fossils of the Shara Miirun beds 


A trip to Jisu Honguer ..... 

. 203 

Ula Usu to Ardyn Obo ..... 

• 203 

Fossil fatina at Ardyn Obo ..... 

. 204 

The Oasis of Gatiin Bologai ..... 


A visit to our caravan ..... 

. 205 

On short rations ...... 

. aos 



Westward from Ardyn Obo ..... 

. 206 

Fourth of July, 1923 ...... 


Compulsory military service in Mongolia . 

. 207 

Camp at the Flaming Cliffs of Shabarakh Usu 

. 307 

Dinosaiu- eggs ....... 

. 208 

The egg-stealing dinosaur ..... 

. 208 

The preservation of dinosaur eggs . . . . . 


A dinosaur nesting place ...... 


A parrot-beaked fossil, Protoceratops andrewsi 


Some birdlike dinosaurs of Cretaceous age 

• 213 

A fossil crocodile, Shamosuchus . . . . . 

. 314 

Hunting bighorn sheep and ibex on Artsa Bogdo 

. 214 

Mongol ponies ....... 

• 3IS 

A search for our caravan ...... 

. 216 

The Gashato formation and its fauna . . . . 

. 317 

Method of packing fossils ...... 

. 3l8 

A huge herd of antelopes near Gurbun Saikhan Mountains 


On leaving the Flaming Cliffs . . . . . 

. 219 


• 331 

The Oshih basin ....... 

. 321 

A pre-Mongol wall ....... 

. 331 

Badlands ........ 

. 331 

The "cannon-ball" sandstone . . . . . 

. 333 

Fossil fauna at Oshih ...... 

. 333 

Living favma at Oshih ...... 

. 333 

Hunting trip to Artsa Bogdo . . . . . 

. 323 



Oshih to Ula Usu ...... 

Arrangements for Professor Osbom to visit the Expedition 

News of the Japanese earthquake 

Professor Osbom 's first visit to Mongolia . 

The second Coryphodont tooth 

Wolves and antelopes 

Making camp in fourteen minutes . 

Plans for the future .... 

Commerce on the heels of scientific exploration 
Plans for a Chinese National Museum 
End of the 1923 Expedition . 


Popular interest and support 
Study of the 192 2-1923 collections 
A new fleet of cars 

Widening the scope of Expedition activities 
A trip to Urga to obtain government permits 
A summer of unusual rains and floods, 1924 
Negotiations in Urga .... 
General Kozloff, a Russian of the old school 
Acknowledgment to T. A. Badmajapoff 
Supplies and equipment for 1925 


The need for a more accurate map . 
Sickness of one of our staff at Kalgan 
Start of the main party 
Activities of bandits . 
Changpeh to Wolf Creek 
Expedition activities at Wolf Creek 
Increased acreage vmder cultivation 
Mvmitions for Marshal Feng Yu-hsiang 
At Tserinville .... 
Method of topographic surveying . 
Detention of the caravan 
Dealing with Buriat officials 














23 s 







Activities at Ula Usu 
Baiying Bologai 
Gatun Bologai 

Trees as evidence of climatic change 
Baiying Goshigo 

Coordination of activities while traveling 
A day of hard work on the trail 
Khundelungi Usu 
Seventy-two miles of arid gravel 
Jichi Ola .... 

The Gurbun Saikhan 

Speed and endurance of a wounded antelope 
Return to the Flaming Cliffs 
Dime Dweller artifacts 
Practical jokes 

Palaeontologic and topographic work 
More dinosaur eggs . 
A nesting site for giant ostriches 
Severe sandstorms 
A necessary trip to Urga 
A May snowstorm 
Arrival in Urga 
Friends in Urga 
Uncertainty of Urga "laws" 
The demands of the Mongol government 
A second meeting with General Kozloff 
The Urga government permits received 
Scientific Committee representatives 


The return from Urga to Shabarakh Usu . 

Origin of the Shabarakh Usu basin 

The primitive culture of the "Dune Dwellers" 

A Mongol cache ...... 









Botanical work 

Mammal skulls from the Age of Reptiles 

Significance of these archaic mammal skulls 

Shabarakh Usu to Tsagan Nor 
Tsagan Nor, a disappointment 
The rise and fall of Tsagan Nor 
Sudden weather changes 
Motion pictiares of wild asses 
A Baluchttheritwi that died in quicksand . 
"Paper-shales" at Moimt Uskuk 
Ascent of Baga Bogdo 

More fossils from the Himg Kureh formation 
The extinct giant ostrich, Struthiolithus 
Structure and surface features of Baga Bogdo 
Trees in the canyons of the Altai Mountains and 
Pre-Mongol graves near Baga Bogdo 


Exploring the Orok Nor district 

Discovery of the Kholobolchi Nor 

Camp at Kholobolchi Nor 

Oligocene and Pleistocene fossils 

Dune Dwellers and other former inhabitants 

Waterfowl and lake plants at Orok Nor . 

Glacial cirques on Ikhe Bogdo 

Problem of the loess deposits 

The land beyond the Altai . 

A mirage that was mapped . 

"Kozloff Pass" 

"Deserted Valley" . 

On the southern side of Ikhe Bogdo 

The Great Mongolian Road to Turkestan 

Wild camels of the Black Gobi 

Reconnaissance along the Great Mongolian Road 

Poor country for fossil-hunting 

Gobi Desert 










A peculiar malady .... 
Foitrth of July, 1925 .... 
The fish of the Tsagan Nor region . 
Bats in the Gobi .... 


Reconnaissance westward from Kholobolchi Nor 

The Baidarik Gol . 

To the Sassaktu Khan Mountains and return 

Side trips from Kholobolchi Nor 

The disappearance of Tsagan Nor . 

Completeness of Buriat domination 

A Dune Dweller jasper quarry near Shabarakh Usu 

Reconnaissance south of the Gurbim Saikhan 

The first argali shot from an automobile . 

Eastward on the south side of the Gurbun Sailchan 

A rumor of war in China 

A "zero hour" windstorm 

The Expedition reassembled at Shabarakh Usu 

"Bigger and better eggs" 

Good-bye to Shabarakh Usu 

A second accident at Clutch Camp 
Hunting bighorn sheep by automobile 
Observations at Golobai-in-Ola 
' Chromite ore at Gutul Usu . 
Experiences at a border yamen 
Return to Inner Mongolia . 
Plans for closing the 1925 season 
Dr. Chaney's summary of his botanical work 
Photographing a great herd of antelope 
A trip to Peking .... 

Return from Peking to Baron Sog . 
New titanothere fossils near Baron Sog 
Ancient human graves near Baron Sog 
An old, abandoned copper mine near Ula Usu 









A visit from the great antelope herd 

Lama superstitions at Baron Sog-in-Sumu 

Viper Camp ..... 

Hoofed animals of Eocene time 

Reconnaissance trips north and west from Viper 

End of the 1925 Expedition . 

Study of the Cretaceous period in Asia 

Winter activities of the staff 
A fossil redwood forest in Manchuria 
Preparation of fossils in China 
The staff for 1926 .... 
Outbreak of civil war in China 
Mo\'ing supplies from Tientsin to Peking . 
Warlike activities delay movement of supplies 
Return of the Leader to Peking 
Getting the equipment to Kalgan 
Air raids on Peking .... 
Narrow escapes from Chinese bullets 
Besieged in Peking .... 
Arrival of Mr. and Mrs. Burden 
Failure of efforts to reach Kalgan 
Return of Young from Kalgan to Peking . 
The Jehol-Dolon Nor route . 
An unpleasant experience with illiterate soldiers 
Abandonment of the 1926 Expedition 
Kalgan affairs and the caravan 
Work in Yunnan and Szechwan 
Lectures in America and in England 

Rise of the anti-foreign movement in China 
Disappointing results of work in Yunnan . 
Tense situation in Peking 
A raid on the Soviet Embassy 
Impossibility of a 1927 expedition . 










The Sven Hedin expedition . 

Plans for 1928 

Brigandage beyond Kalgan . 

Exorbitant taxes and graft . 

Cooperation by the press 

Agreement with the brigand representative 

Staff of the 1928 Expedition . 

Objectives for 1928 . . . 

Diplomatic influence . 
A bandit ruse that failed 
The Swedish mission of Mr. Eriksson 
Mongol diseases and medicines 
A visit to Prince West Sunit 
Dune Dweller artifacts near Hatt-in-Sumu 
Hatt-in-Simiu to Chimney Butte 
Work at Chimney Butte 
Victrola and radio 
A spell of bad weather 
Vague news from the Sven Hedin expedition 


An exploring trip westward from Chimney Butte 

The return to Chimney Butte 

Decision to follow Sven Hedin 's route 

Further work at Chimney Butte 

Injury to the Leader . 

New formations and fossils east of the Shara Murun 

Severe weather in mid-May . 

Exploring trip to Iren Dabasu 

War news from China 

Relations with local Mongols 

The region south of Baron Sog monastery 
Chinese chauileurs .... 








"Land of the Larks" 


Failing Miao ...... 


The "Winding Road" to Turkestan 


The provincial boundary between Chahar and Suiyuan 


News from the Sven Hedin expedition 


A Chinese merchant caravan 


Opium smugglers ..... 


Shirigi-in-Sumu, a deserted monastery 


Activities while waiting for the caravan 


A sandstorm for a day and a night . 


Stopped by Mongol soldiers .... 


In camp near Dime Dweller stations 


The Lang Shan Mountains .... 


Desolate and arid coimtry .... 


Junction of the Winding Road and the Small Road 


Impossible conditions for plane-table mapping 


The Laohu Shan, or Tiger Mountains 


A disappointing sedimentary basin . 


News of Chinese political affairs 


Go Yoto basin, the "Very Bad Place" 


The end of the trail for us . 


Abandonment of the western trip 




Serious condition of the camels 


Hunting argali ..... 


Peas for our caravan ..... 


Tax harriers ...... 


The Sha Kang Usu escarpment 


A Dune Dweller station .... 


Relaying supplies by automobile 


Reconnaissance ea.stward from Shara Murun 


At Erhlien 


Eastward from Erhlien .... 


The Gur Tung Khara basin .... 


The Baiying Obo plain .... 




The Ulan Hsanda basin 
Tsagan Cholo hills 
On the Dolon Nor trail 
A disabled car . 
Return to Erhlien 
News from China 
Collecting at Urtyn Obo 


Fossil-hunting with field-glasses 
Conflict with lama superstitions 
Baluchitherium bones . 
Bones as big as Shackelford's body 
Other fossils at Urtyn Obo . 

Various activities at Urtjm Obo 
Dime Dweller artifacts 
Animal pets 
Obnoxious lamas 
A spell of terrific heat 
Caravan affairs 

Nom Khong Shireh, the Holy Mesa 
Serious loss of gasoline 
Operations at the Holy Mesa 
News from China at Iren Dabasu 
Three sets of dinosaur eggs . 
Iren Dabasu to Our Tung IChara Usu 
Pliocene lake beds of the Timg Gur forma 
Recurring rainstorms . 
A trip to Hatt-in-Sumu and return 
Problem of gasoline supply . 

Camp at Tairtun Nor 
Fossiliferous exposures at Tairum Nor 
An ancient grave 









The shovel-tusked mastodon, Platyhelodon 

Dune Dweller artifacts between Tairum Nor and the Dolon Nor trail 

Traveling in rain and cold weather . 

Hailstones an inch in diameter 

The Barro Unduh upland . . . 

Koko Nor ..... 

Village of East Sunit Wang, at Unduh Cholo 

Attempt to reach Hul Tsagan Nor . 

Grave of the former prince of East Sunit . 

Gosho-in-Sumu .... 

Heavy sand west of Gosho-in-Sumu 

Artifacts with bone, at Chilian Hotogha . 

Problem of the Dune Dwellers 
The sand-dune area north of Dolon Nor . 
Return to Hatt-in-Sumu 
Side trip to Gul Chagan 
Final caravan arrangements . 
Hatt-in-Sumu to Wan Ch'uan Pass 
Injury to Horvath .... 
Arrival in Kalgan .... 
Kalgan to Peking by automobile 

Trouble with the Society for the Preser\-ation of Cultural Objects 
The anti-foreign movement . 
End of the 1928 Expedition . 


Conditions in Peking, March, 1929 

Demands of the Commission for the Preser\'ation 

Agreement for a 1930 Expedition . 

The 1930 stafi ..... 

Experiences with Chinese Customs officials 

Changpeh to Hatt-in-Sumu . 

News of Pliocene fossils at Tulchum Nor . 

Pleistocene bones near Hatt-in-Sumu 

of Ancient Objects 














A duststorm at Hatt-in-Sumu 

Unsuccessful attempt at eastward reconnaissance 

Through the sand-dunes to Tukhum Nor . 

Disappointing nature of the Tukhum Nor exposiu-es 

Shelldrake burrows in P'ang Kiang formation 

Fossil beds near Honger Obo 

Abandonment of the sand-dune area 

Plans for further work at Mastodon Camp, 1930 
Repairs after a duststorm 
Effects of rain in the desert . 
Through and beyond P'ang Kiang . 
Bulga-in-Sumu . ... 

Again at Mastodon Camp 
Exploration of an ancient lake-shore 

Wolf Camp 

Water-supply at Wolf Camp 

A mother and baby shovel-tusked mastodon 

Arrival of the caravan at Wolf Camp 

Other fossils at Wolf Camp . 

Reconnaissance east of Wolf Camp 

A death-trap for shovel-tusked mastodons 

Bones from the crowded grave 

XLII.— WOLF CAMP ...... 

Routine work ..... 

The fresh meat supply 

Lost on the plains when hunting wolves . 

The abundance and activities of wolves 

A baby gazelle as a camp pet 

A violent storm .... 

Loss of the alcohol supply 
A trip to Peking for alcohol . 
Return from Peking to Wolf Camp 
More new fossil types 









An unborn baby mastodon . 

Transport of fossils to Peking 

A large herd of gazelles 

Fauna of the Pliocene beds at Wolf Camp 

Chinese opposition to further work 

Possibihty of finding remains of early man 

End of work at Wolf Camp . . 


Camp Margetts .... 

Fossil-bearing exposures near Camp Margetts 

Animal life in the badlands . 

Nightmare creatures of the past 

Transport of more fossils to Peking 

Strange new fossil amblypods 

Fossil turtles ..... 

Significance of a charcoal deposit 

Other early Tertiary fossils . 

More promising exposures 

Final return of the Leader to Peking 

Young's battle with brigands 

Brigands as "road police" 

Close of the 1930 field season 

Significance of the fossil collections 

The unfinished task .... 



By Roy Chapm.^n Andrews 

Our quest begins 

From Peking to Kwanyingtang by rail 

Travel by mule Utter . 

On the road to Sianfu 

Trafific on a Chinese road 

Tungkwan .... 

A side trip to Hua Shan— The Flower Mountain village 










The ancient capital, Sianfu . . . 

Military disturbances 

Lingtai Miao — Hunters' headquarters 

Camping in a temple at the foothills of the Taipai Shan 

Engaging a hunter ..... 

We set out for the hunt .... 

A mountain meadow camp .... 

Liu predicts success ..... 

Climbing to the peaks of the Taipai Shan 
Takin at last ...... 

Stalking the game in the bamboo jimgles . 

A wet night in the gorges .... 

My first takin ...... 

Collecting background material for Museum group 
Htmting near snow-line .... 

We leave our hunters to complete the work 

Our contact with retreating soldiers 

We keep hospital hours .... 

Our return to Peking ..... 


By Clifford H. Pope 

My first collecting trip in China 

Reptiles and amphibians of the Ttmg Ling region 

Collecting at Kichowpeh 

The "art" of collecting in China 

A trip to the Yangtze valley in Anhwei Province 

Collecting methods at Ningkwofu . 

Snakes of the Ningkwofu area 

Tiutles, frogs and a salamander 

Fish fauna near Ningkwofu . 

Trip to Tungting Lake, Himan Province . 

A rare river dolphin .... 

Mammals of northern Hunan 

Fish faima of Tungting Lake 

Cooperation of Huping College faculty 
















Alligators from Wuhu in the Yangtze valley 
Alligator dens ..... 
Transporting dormant alligators 
Survival of alligators in a populous region 
Relation of Chinese and American alligators 
Suiyuan Province .... 
Fauna of Shansi Province 
Value of a knowledge of the Chinese language 

-ISLAND OF HAINAN, 1922-1923 

By Clifford H. Pope 

Journey to Hainan 

Cooperation of missionaries at Nodoa 

Difficulties of work in Hainan 

Ah-sen, guide and interpreter 

Methods of collecting fish 

Activities at Nodoa . 

Bargaining with Hainanese farmers 

Problems of artist Wong 

Indifference of Hainanese to sim and rain 

Activities of outlaws . 

Fish, amphibians and reptiles of Hainan 

Mammalian fatma 

End of work in Hainan 


By Clifford H. Pope 

Importance of Fukien for herpetological collecting 

Native helpers .... 

Assistance of missionaries 

Collecting near Yenping 

Breeding of the tree-frog, Polypedates dennysi 

Kuatun, a famous collecting locality 

An isolated Roman Catholic community . 

Return from Kuatun to Foochow . 

Futsing Hsien and the Lingshihszu Moimtains 










Reptiles and amphibians of the Lingshihszu Mountains 

Mammals of the Lingshihszu region 

Tigers ..... 

Packing specimens for shipment 

Second trip to northwestern Fukien 

Collecting mammals at San Chiang 

Cooperation of native assistants 

Children as collectors 

Larvae of a primitive salamander 

A snake as a collector 

Adaptation of frogs to cascading streams 

Summary of Kuatun fauna . 

Fish fauna of Ch'ungan Hsien 

Small collection from Hok'ou, eastern Kiangsi 

Close of work in Fukien 



Winter Seasons of 1921-1922, 1922-1923, and 1925-1926 
By Walter Granger 

My arrival at Peking .... 

Fossils reported on the upper Yangtze River 

Selecting a field staff ..... 

Peking to Hankow ..... 

On board the "Tung Wo," a Yangtze River boat 

Low-water and high-water levels on the Yangtze River 

Inter-provincial warfare along the Yangtze 

On board the "Limg Mow" at Ichang 

Through the Yangtze Gorges, Ichang to Wanhsien 

At Wanhsien ...... 

Fossil bones and their uses .... 

A trip to Yen-ching-kou, a fossil locaUty . 

Living quarters, at Yen-ching-kou . 

The mode of occurrence of fossils at Yen-ching-kou 

Chinese methods of collecting fossils 

Oxu- methods of collecting .... 

Pliocene and Pleistocene faunas 








Evidence of contemporary human occupation 
Other fossil localities near Yen-ching-kou . 
Relations with the villagers and farmers 
Preparations for departure .... 
By junk from Wanhsien to Ichang . 
From Ichang to Peking .... 

Succeeding seasons at Yen-ching-kou 
Season of 1922-1923 ..... 
Season of 1925-1926 ..... 
Inter-provincial warfare, 1922-1923 and 1925-1926 
Collecting during the second and third seasons 
Results of our work at Yen-ching-kou 


By Wai.ter Granger 
Why the Yunnan trip ? . . . . 

Down the China coast by boat from Tientsin to Haiphong 
Haiphong to Yunnanfu .... 

Headquarters at Yunnanfu .... 
The Red River trip ..... 
Collecting at Kao Chiao Temple 
The illness and passing of Hui 
A trail in mountainous country 
A soldier escort ..... 

The serious illness of Chow .... 
Yuan Kiang on the Red River 
At the Mission, near Yuan Kiang . 
The return journey to Yimnanfu 
The Ta-hu-kai shell heap on the shore of Kun-yang-hai Lake 
Results of the Red River trip 
Outfitting- at Yimnanfu for the northern trip 
The trip to the Yangtze River 
At the Miao Inland Mission at Sapushan . 
Foreign tribes in southern China 
Bandit-infested country beyond Wu Ting . 
In the Ma Kai valley .... 









At the city of Ma Kai .... 

At Lung Kai on the Yangtze River 
Nelson finds a Neolithic culture site at Lung Kai 
Early Pleistocene fauna in Ma Kai ceolian deposits 
Bandits prevent exploration of Amichow caves . 
The principal results of the trip to Yunnan 


By Nels C. Nelson 
Preliminary activities 
The Expedition under way . 
Equipment and personnel 
Work within the Gorges 
The discovery of Neolithic culture sites 
Conclusions drawn 



By Chester A. Reeds 
Introduction ..... 
Exploration in Mongolia and China 
Preliminary and final reports 
The geographical features of Mongolia 
Weather changes in Mongolia and China . 
Topographic surveys .... 
Rock formations in the Gobi region 
The oldrock complex .... 
The "later sediments" 
Mongolian erosion planes 

Changes in climate in Mongolia during geologic periods 
Fossil faunas and horizons . 
The living and fossil flora 
The living faunas of Mongolia and China . 
Prehistoric archaeology 
Conclusion ..... 











By Various Authors 


By Chester A. Reeds and Charles P. Bereey 

1. Unsolved Geologic Problems . 

By C. p. Berkey, A. W. Grabau, F. K. Morris and L. E. Spock 

Locality studies ...... 

Stratigraphic succession and age relations . 

Subdivision of the Pre-Cambrian 

Age and correlation of the Khangai Graywacke Series 

Pre-Permian stratigraphy 

Post-Permian stratigraphy 
Deformation and structure 

Petrographic problems 
Physiographic problems 
Glacial history and climate 
Prehistoric archaeology 
Mineral resources 

2. Unsolved Problems in Vertebrate Palaeontology . 

By Henry Fairfield Osborn and W^u-ter Granger 

3. Problems in Mammalogy ..... 

By Glover M. Allen 

4. Unsolved Problems in Chinese Ichthyology 

By J. T. Nichols 

5. Archaeological Studies and Problems in Central Asia 

By Nels C. Nelson 

6. Problems in Palaeobotany in Northeastern Asia . 

By Ralph W. Chaney 

7. Problems in Herpetology ..... 

By Clifford H. Pope 

8. A Proposed International Institution for Asiatic Research 

By Roy Chap^lan Andrews 

Final Generalization of Resiilts 


Bibliography ..... 

Publications of the Central Asiatic Expeditions 
Contributors to the Central Asiatic Expeditions 
Index ...... 






58 r 












I. — Professor Osbom's Prophetic World Map of 1899-1900, Science, page 567, April 

13. 1900 3 

2. — Map of Mongolia superimposed upon map of United States to show comparative 

size .............. 6 

3. — Generalized geologic section from Tientsin to the Wan Ch'uan Hsien Pass. From 

"Geology of Mongolia." .......... 25 

4. — Two cross profiles of the Gobi Basin. From "Geology of Mongolia" . . . 26 

Profile A crosses the whole of the eastern Gobi and shows a broad, shallow down 
warp between the Arctic and Pacific Divides. 

Profile B crosses only the northwestern extension of the Gobi lying north of the 
Altais and shows much stronger warping and block-faulting than does the eastern 

S • — Geologic section through the low rolling coimtry north of the Chahar Mountains. 

From "Geology of Mongolia" ......... 35 

6. — Entering the P'ang Kiang Basin — a typical sediment basin of the "Gobi." From 

"Geology of Mongolia" .......... 35 

7. — The scarp at Irdin Manha. From "Geology of Mongolia". .... 38 

8. — Margin of a roof-pendant north of Tsetsenwang. From "Geology of Mongolia". . 74 

9. — Drawings cut upon a basalt block at Artsa Bogdo. From "Geology of Mongolia". 151 

10. — Generalized columnar section of the oldrock floor. From "Geology of Mongolia". 561 

1 1 . — Summary of Mongolian formations, faimas and geologic history from the Jurassic 

up. After Osbom ........... 566 

1 2 . — The later sediments of Mongolia with some representative forms from each faima. 

After Osbom ............ 567 


Relief map of Mongolia showing routes, Central Asiatic Expeditions, 1922-1930 . at etid 

Route map of the Central Asiatic Expeditions, 1922-1930 ..... at end 

Map of China showing in red location of places visited by members of the Central 

Asiatic Expeditions, 1921-1930 ........ at end 


Colored Frontispiece: A Mongol woman of Urga with a lama and prayer wheel. From a 
painting by Arthur A. Jansson. 


I. — Staff of 1922 Expedition in the office of the Peking Headquarters. Left 
to right : Colgate, Wong, Pope, Andrews, Berkey, Morris, Granger 
and Shackelford ......... 4 

II. — 1922 Expedition at Tsagan Nor, Mongolia. Bottom row: Mongol inter- 
preters and caravan men. Second row, left to right : Morris, Colgate, 
Granger, Badmajapoff, Andrews, Berkey, Larsen, Shackelford. Top 
row : Chinese technical and camp assistants . . . . . 5 

III. — Staff of 1923 Expedition at Irdin Manha. Bottom row: Chinese technical 
assistants and Mongol interpreters. Second row, left to right : Gran- 
ger, Osbom, Andrews, Morris, Kaisen. Top row, second from left: 
Vance Johnson, Albert Johnson, Young, Olsen . . . .12 

IV. — 1925 Expedition at Shabarakh Usu. Seated, left to right: Roberts, 
Andrews, Granger, Morris, Berkey, Young. Standing: Robinson, 
Olsen, Chaney, Butler, Nelson, Loucks, Lovell, Shackelford . . 13 

V. — Staff of 1928 Expedition at Hatt-in-Sumu. Seated, left to right: Perez, 
Andrews, Granger, Spock, Thomson. Standing: Shackelford, Pond, 
Eriksson, Horvath, Young, Hill . . . . . . .16 

VI. — New members of the staff of the 1930 Expedition . . . • 17 

(A) Pere Teilhard de Chardin, Geologist. 

(B) Lieutenant W. G. Wyman, U. S. A., Topographer. 

(C) Dr. A. Z. Garber, Surgeon. 

VII. — Headquarters of Central Asiatic Expeditions at 2 Kung Hsien Hutimg, 
Peking, China. Equipment and stable court, outer laboratory and 
office, entrance to inner court, servants' quarters .... 20 

VIII. — Kalgan from outside the Great Wall looking south .... 28 
The main gate may be seen in the center of the picture. At the right 
the Great Wall of China defending the Pass climbs over the mountains. 

IX. — (A) Typical page from geologists' daily route notebook . . -32 

Professors Berkey and Morris carried their geological sections in this 
way throughout the entire route of the Expedition. 
(B) An Expedition car in difficulties, 1928 ...... 32 

Often an apparently hard crust would break, letting the car sink into 
soft, sticky mud. 


xlii PLATES 


X. — Camels being watered at a desert well ...... 33 

It was from such wells that the Expedition usually obtained its water- 

XI. — The telegraph station at Iren Dabasu (Erhlien) ..... 40 

XII. — (A) Tserin, Mongol leader of the Expedition's caravan, 1928 41 

(B) A prayer wheel and a lama at Urga ...... 41 

(C) A five-year-old Mongol sheep herder . . . -41 

(D) Sewing a patch of leather over an injury on a camel's foot 41 

XIII. — (A) A pre-Mongol grave at Tsetsenwang, 1922 . . 52 

These graves exist in many places in the grasslands of Outer Mongolia. 
(B) A pre-Mongol skeleton in a grave at Tairum Nor, 1928 ... 52 
The remains of a skirt composed of bits of fossil shells may be seen about 
the middle of the skeleton. 

XIV. — (A) Exterior view of the prison at Urga ...... 53 

(B) Prisoner confined in a coffin ....... 53 

Fortunately this barbarous method of punishment is no longer used in 

XV. — (A) Mongol women of the Khalka tribe at Urga ..... 60 
This is the typical headdress of the women of northern Mongolia. 
(B) The Great Temple in the Lama City at Urga .... 60 

XVI. — (A) A Mongol giant at Urga 61 

This man was nearly eight feet tall and weighed 307 pounds. 
(B) The Maidari procession in Urga, 1922 ...... 61 

The Expedition made the first motion pictures of this festival. It has 
now been abolished. 

XVII. — (A) A pet red -billed chough tells the Chief Palaeontologist a secret. . 72 

(B) Branding Mongol ponies in Mr. Larsen's compound at Urga, 1922. 72 

XVIII. — Mongols listening to a phonograph ....... 73 

Usually they were greatly frightened when hearing it for the first time. 

XIX. — The Expedition camp in the forest north of Sain Noin Khan, 1922 . 84 

XX. — The temples and Lama City at Sain Noin Khan, 1922. ... 85 

XXI. — (A) Yaks in the forest region of northern Mongolia .... 92 
(B) Prince Regent at Sain Noin Khan ...... 92 

XXII. — Camel and motor car tires, Shabarakh Usu . ..... 93 

XXIII. — Mongol village near Artsa Bogdo 96 

The village was situated on the edge of a meadow-land watered by 
streams flowing down from the mountain. 1925. 

XXIV. — (A) Mongols gathering salt 97 

The suri'ace of this lake had a crust of pure salt an inch thick, in 1922. 
(B) Mongols inspecting flashlights . . . . . . -97 

XXV. — Camels in the sand-dunes near Tsagan Nor, 192s. .... 112 



PLATES xliii 


XXVI. — (A) Part of a herd of wild asses near Tsagan Nor .... 113 
(B) Wild ass running at a speed of forty miles an hour, 1925. . . 113 

XXVII. — Camels in the sand-dunes between Tsagan Nor and Baga Bogdo, 1925 . 120 

XXVIII. — Pushing one of the Expedition cars through sand . . . .122 

Strips of canvas, reinforced with transverse pieces of rope, were often 
placed in front of the wheels. 

XXIX. — (A) Wild ass running at full speed, Tsagan Nor, 1925 . . . .123 

(B) Head of goitered gazelle, Procapra gutturosa . . . . .123 

(C) Head of desert gazelle, Gazella subgutturosa hillieriana . . .123 

(D) Head of Mongolian black vultiu-e ...... 123 

XXX. — (A) Framework of a yurt as it is being erected . . . . .128 

(B) Yurt and corral for lambs, kids and calves ..... 128 

XXXI. — (A) Women of the Chahar district, southern Mongolia 

(B) Interior of a yiirt ......... 

XXXII. — (A) Mongols making felt, Tsagan Nor, 1922 

(B) Mongols milking goats. The goats are held in Une by a long rope 

which is looped arotmd the neck of each animal . . . .132 

XXXIII. — (A) Andrews dressing an injury to a Mongol girl, Tsagan Nor, 1922 . 133 
(B) Typical Mongol nomads, Shabarakh Usu, 1925 .... 133 

XXXIV. — Restoration of Baluchitherium under the direction of Henry Fairfield 

Osbom. Painted by Charles R. Knight, 1923 .... 140 

XXXV. — (A) Halt for tiffin in the sand-dunes south of Tsagan Nor, 1922 . . 141 

(B) Skull of Baluchitherium discovered in 1922 ..... 141 

XXXVI. — Camels in the sand-dunes near Tsagan Nor, 1925. .... 144 

XXXVII. — (A) Granger removing a fossil deer antler. Foot of Baga Bogdo, 1922 . 145 
(B) Granger inspecting a bush of sweet pea. Sand-dunes near Tsagan Nor 145 

XXXVIII. — (A) Ibex coiuitry on Artsa Bogdo, 1923 ...... 150 

(B) Mongolian ibex, Capra siberica, 1923 . . . ... . 150 

XXXIX. — (A) A valley on the slopes of Artsa Bogdo, 1923 . .... 151 
(B) The "Shirto House," a cave on a summit of one of the peaks of Artsa 

Bogdo, 1923 151 

XL. — Tserin points out approaching caravan to Andrews, Chimney Butte, 1928. 154 

XLI. — (A) Ancient wall at the base of the Red Mesa, Oshih Basin, 1923 . . 155 

(B) Expedition camp in the Oshih Basin, 1923 . . . . ■ iSS 

XLII. — A detail of the Flaming Cliffs, showing wind erosion .... 162 

XLIII. — (A) "The dinosaur," a natural wind-carved red sandstone remnant. 

Cretaceous beds of Shabarakh Usu . . . . . .163 

(B) A fossil deposit at Ardyn Obo, 1922 ...... 163 

XLIV. — The Flaming Cliffs at Shabarakh Usu where the dinosaur eggs were dis- 
covered, 1925 .......... 168 

XLV. — Mongol herders carrying lassoes at a desert well . . . . 174 

xliv PLATES 


XLVI. — (A) Frederick Morris at the plane-table, Oshih Basin, 1923 . . -175 

(B) McKenzie Young with a mountain sheep head. Artsa Bogdo, 1923 175 

(C) Professor Henry Fairfield Osbom holding the newly found duplicate 
tooth of the coryphodont, Eudinoceras mongoliensis. Irdin Manha, 

1923 175 

(D) J. B. Shackelford with two marmots in spring pelage. Five Antelope 
Camp, 1922 . . 175 

XLVII. — Artsa Bogdo, showing the main peak, Tiger Canyon and part of an 

alluvial fan .......... 184 

XLVIII. — (A) Skull of Andrewsarchus and of an American timber wolf. . . 188 
(B) Restoration of Andren'sarchus under the direction of Henry Fairfield 

Osbom, by Mrs. E. R. Fulda 188 

XLIX. — (A) Dinosaurs at Iren Dabasu, painted by Mrs. E. R. Fulda 189 

(B) Entelodon, a giant pig-like animal, painted by Charles R. Knight . 189 

L. — (A) Restoration of Protoceratops on the nesting grounds of Shabarakh 

Usu during the Cretaceous period, by Mrs. E. R. Fulda. . 194 

(B) Skeleton of Protoceratops andren'si from Shabarakh Usu, mounted by 

Peter Kaisen, American Museum of Natural History 194 

LI. — (A) Sloill of Protoceratops andrewsi from Shabarakh Usu, 1923 195 

(B) Skull of Protoceratops andrewsi partly excavated at Shabarakh Usu, 

1923 195 

LIL — The first nest of dinosaur eggs discovered by George Olsen at Shabarakh 

Usu in 1923 .......... 206 

Two eggs and part of another are shown lying on the sxirface. The 
small sandstone ledge in the background was removed intact and sent 
to the Museum. In the center of the block of stone thirteen other eggs 
were discovered, 1923. 

LHL — (A) Packing fossils at Shabarakh Usu, 1923 ..... 207 

(B) Skeleton of Oviraptor, a small species of toothless dinosaur. 207 

This specimen was found lying directly above the first nest of dinosaur 
eggs discovered in 1923. Presumably it was an egg thief. 

LIV. — Andrews and Tserin at Hatt-in-Sumu, 1928 . . . . . .210 

LV. — (A) Bighorn sheep of the Altai Mountains at Artsa Bogdo . . .211 
(B) Head of Oot's awwoM . . . . . . . .211 

LVI. — (A) Professor Osbom and Walter Granger at Irdin Manha, 1923 . .222 

(B) Professor Osbom and McKenzie Young at the barrier, Kalgan, 1923 222 

LVII. — (A) Professor Osbom, Chief Andrews, Assistant Chief Granger. Sep- 
tember 16, 1923. ......... 223 

Irdin Manha horizon. Pit which yielded the first perfect titanothere 
skull, in the month of May, 1923. 
(B) Mongol camel riders with lassoes. ...... 223 

LVIII. — Norman Lovell repairing a motor car, 1925. ..... 232 




LIX. — A commercial transport car running between Urga and Kalgan 

This five-passenger Dodge motor car carried a load of 1,400 pounds, 

LX. — Crossing the Ongin River ......... 

Canvas hoods are placed over the front of the cars. 

LXI. — (A) Roberts, Butler and Robinson taking observations of the sun. 
Shabaraldi Usu, 1925. ........ 

(B) The Expedition staff at luncheon, Shabarakh Usu, 1925. 

LXII. — Granger removing a nest of dinosaur eggs at the Flaming Cliffs, 1925 

LXIII. — Andrews and Olsen with twelve dinosavu" eggs at Shabarakh Usu, 1925 

LXIV. — A dozen dinosaur eggs discovered by George Olsen at Shabarakh Usu, 1925 

LXV. — (A) Five dinosaur eggs ....... 

This set is in the Field Museum of Natural History, Chicago. 
(B) Restoration of young dinosaurs emerging from eggs . 

LXVI. — Expedition caravan arriving at Flaming Cliffs, 1925 

LXVII. — (A) N. C. Nelson sorting Dune Dweller implements collected at Sha 
barakh Usu, 1925 ....... 

(B) Types of implements m.ade by Dune Dwellers 

LXVIII. — Expedition caravan leaving Shabarakh Usu, 1925 

LXIX. — (A) A snow-covered pass south of Urga, 1925 

(B) Andrews and Young in a snowdrift on the way to Urga, 1925 

LXX. — The eroded area of dead sand-dunes at Shabarakh Usu where remains of 
the Dune Dweller culture were discovered, 1925 .... 
The Flaming Cliffs are in the center distance. 

LXXI. — Looking south toward the main peak of Baga Bogdo from the mouth of 
Tiger Canyon, 1925 . 


LXXn. ^Looking north from Tiger Canyon toward Tsagan Nor. 

LXXIIL— (A) Collecting fish from Tsagan Nor, 1922 . 

(B) Andrews in the mess tent at Kholobolchi Nor, 1925 

LXXIV. — (A) Desert vegetation north of Tsagan Nor 

(B) So-called tamarisk tree ..... 
The base has already been partly buried by drifting sand. 

LXXV. — The Expedition caravan arriving at Camp Chimney Butte, 1928 

LXXVL — Caravan leaving camp at Chimney Butte, 1928 . 

LXXVIL— (A) Doctor Chaney collecting botanical specimens, 1925 

(B) Typical Mongolian oxcart ...... 

LXXVIIL — Andrews, Granger, Young and Thomson repacking supplies at Hatt-in- 
Sumu, 1928 .......... 





















LXXIX.— (A) A Mongol lama 

(B) Red lama 

(C) Mongol woman of the Khalka tribe near Urga 

(D) A Chahar Mongol woman of the southern grasslands 

LXXX. — The Expedition at Goptchil lamasery, 1928. 

LXXXI. — Expedition cars at the Peking headquarters, 1928 

LXXXII. — Starting on the 1928 Expedition. .... 

The American Minister, Mr. J. V. A. MacMurray, accompanied the 
Expedition to the top of the Pass. The cavalry was sent as an escort 
by the Chinese Government. 

LXXXIII. — The Expedition in camp at night listening to the victrola 

Left to right: Hill, Pond, Thomson, Andrews, Granger, Horvath, 
Young, Spock, Perez. 

LXXXIV. — (A) A halt for luncheon on the way to Ula Usu, 1928 . 
(B) Car badly mired, 1928 ..... 

LXXXV.— Camp at Hei Ne Ho, 1928 

LXXXVL — (A) The beginnings of badland erosion, Urtyn Obo, 1928 
(B) A deeply eroded canyon, Urtyn Obo, 1928 

LXXXVn. — Expedition cars at P'ang Kliang, 1928. 

LXXXVHL— Part of badlands at Urtyn Obo, 1928 .... 

LXXXIX. — Excavating the Baluchitherium bones discovered by Mr. Shackelford 
Urtyn Obo, 1928 ....... 

XC. — Removing the bones of a Baluchitherium at the Holy Mesa, 1928 

XCL — (A) Captain Hill surveying at Urtyn Obo . 

(B) Granger excavating the skull of an Emholotiierium 

XCn.— The Obo at the Holy Mesa, 1928 

XCHL — Andrews collecting young kites at Urtyn Obo, 1928 

XCIV.— The badlands at Urtyn Obo .... 

The top layer represents the bright yellow Oligocene sands. Below 
the unconformity the strata are composed of red Eocene sediments 

XCV. — Excavating the skeleton of a rhinoceros at Mastodon Camp, 1928. This 
animal evidently had died in a stream-bed ..... 

XCVL — Filling the Expedition's water-bags at a desert well, 1928 

XCVIL— Dime Dweller site, 1928 

It was here that bones of animals and many flint implements of these 
interesting people were discovered. 

XCVHL — (A) The site of an ancient Dime Dweller hearth 

Remains of fire and of bones were found on the hearth. 
(B) The Expedition's camp servants in front of the cook tent, 1928 















PLATES xlvii 


XCIX. — (A) Side view of the skull of the battering ram titanothere, Emholotherium 424 
(B) Front view of Embolotherium skull ...... 424 

C— (A) The Expedition at "Wolf Camp" 425 

(B) Our pet antelope nursing from its foster mother, a Mongol goat . 425 

CI. — (A) The graveyard of the shovel-tusked mastodons after three weeks' 

excavation .......... 432 

(B) Two jaws of shovel-tusked mastodons from the Great Quarry. . 432 

At the extreme right of the picture may be seen the broad flattened 
tusks of another pair of jaws. 

CII. — (A) Granger and Thomson with a shovel-tusked mastodon's jaw . . 433 
This specimen is still in its field casing of burlap and paste. 
(B) The jaw of a shovel-tusked mastodon resembles an ordinary coal 

scoop 433 

cm. — The battering ram titanothere Embolotherium grangeri. Restoration by 

Margret Flinsch . . . . . . . . .434 

CrV. — The shovel-tusked mastodon Platyhelodon. From a model by Margret 

Flinsch . . . . . . . . . . -435 

CV. — (A) Ova mule litters approaching the walled city of Timgkwan . 460 

(B) Wheelbarrows on the road to Sianfu ...... 460 

The man at the handles, by means of a strap over his shoulders, takes 
much of the weight from his arms, leaving his hands free to guide the 
load; the man in front drags it forward with a rope. 

CVI. — (A) Traffic on a Shensi road . . . . . . . .461 

Two- and three-mule carts, carrier-coolies, mule-litters, ox-trains, camel 
caravans, sedan chairs and wheelbarrows travel these ancient roads. 
(B) Gates of the ancient capital of Sianfu . . . . .461 

CVII. — (A) Camp in a meadow on the Taipai Shan ..... 464 

Tents and skeletons of poles covered with grass and dwarf bamboo 
gave us shelter. 
(B) Chinese hunter with an adult female takin ..... 464 

CVIII. — (A) A bamboo jungle 465 

The dwarf bamboo stalks are not larger than one's finger, and only ten 
to fifteen feet high, but they grow so close together that it is impossible 
to get through without force, or to see more than ten feet ahead. 
(B) A cliff of the Taipai Shan 465 

CIX. — (A) A Chinese cobra. Cobras are common on the Island of Hainan 

where this specimen posed for its photograph .... 484 

(B) The pit-viper, Agkistrodon monticola. It was foxmd guarding its eggs 

in the mountains of northwestern Fukien ..... 484 

(C) C. H. Pope and native collectors. Specimens were frequently brought 
in by crowds of curious farmers who often stayed to watch identi- 
fication and preparation . . . . . . . .484 

xlviii PLATES 


ex. — (A) A Chinese alligatxjr. This torpid alHgator lies bound near its site of 

hibernation a few miles from Wuhu, Anhwei Province . 485 

(B) The Yangtze freshwater dolphin, Lipotes vexillifer. This species was 
not known to science until 19 18. One of the very first specimens 
procured for scientific study is shown here- . . . . 485 

(C) A large pit-vnper of Central China. The mild and docile nature of 
this snake belies its appearance. This specimen was taken in north- 
western Fukien .......... 485 

(D) Hainan Island porcupine. One of the new subspecies collected was 
this porcupine, Acantion subcristatus papcB ..... 485 

CXI. — (A) A Chinese river boat. This boat, like countless others in China, is 
especially adapted to the river for which it is built. We used this 
one on our trip to the mountains of northwestern Fukien . 492 

(B) Kuattm Mountain, near the Kiangsi boundary of northern Fukien. 
It is scientifically famous as the type locality of scores of animals. 
Its summit is approximately 7,000 feet above sea-level, and its flanks 
are well forested . . . . . . . . -492 

CXII. — (A) A view of the top of one of Fukien's highest ranges taken at Upper 

Kuatun. A tea field and bamboo grove appear in the foreground . 493 
(B) The Bohea Hills (Wuishan), Fukien Province. These hills are no less 

famous for their tea than for their scenic beauty. . . . 493 

CXIII. — (A) The bund at Ichang at medium high water. A British gunboat in 

the foreground. September i, 192 1 ...... 502 

(B) Sampans on the Yangtze River. Dr. Granger's party proceeding up 

river from Wanhsien to the landing at Pai-shui-chih . . . 502 

(C) An arched bridge at the mouth of Pai-shui-chih. These lovely 
arched bridges are not xmcommon in this part of China . . .502 

(D) A chartered Chinese junk. Dr. Granger's party proceeding down 

the Yangtze River to Ichang. March, 1923 ..... 502 

CXIV. — Yen-ching-kou. Looking down the valley, toward the Yangtze from near 
the base of the Dragon Bone ridge. The upper village and temple in 
the middle distance. In the foreground an inn known as the "White 
House." The white streak to the right of the paddy fields is the main 
trail between Hupeh Province and the river . . . -503 

CXV. — (A) Yen-ching-kou. The main highway through the lower village. The 

Dragon Bone ridge is seen in the background .... 506 
(B) The 1922-1923 party in the temple at Yen-ching-kou. Foreground 
— Granger, Wong. Middle row, left to right, Chow, Number One 
Boy; Kan (Buckshot), chief technical assistant; Chih, taxidermist; 
Whey, cook. In the background two local coolies, put in imiform 
to insure against drafting by the soldiery .506 

CXVI. — The upper village of Yen-ching-kou with the temple in which the party 

lived diuing three winters . . . . . . . • 5°? 

PLATES xlix 


CXVII. — (A) The temple at Yen-ching-kou. Showing the residence gallery and a 

comer of the stage and sections of the ornate roofs . . .516 

(B) Cultivated fields at the base of the Dragon Bone ridge. At the left 
are the stone steps up which the 1,700-foot climb to the fossil pits 
was usually made . . . . . . . . .516 

CXVIII. — (A) The temple at Yen-ching-kou. Showing the stage and a part of one 

of the long galleries. Christmas time, 1922 . . . . -517 

(B) A pit in the middle of a cultivated field on the Dragon Bone ridge. 
The excavation of this particular pit yielded no fossils — one of the 
nimierous blanks which the excavators draw . . . . -Si? 

CXIX. — (A) A highly fossiliferous pit in working order on the Dragon Bone ridge 524 
(B) An abandoned pit. This was excavated to a depth of fifty feet and 
yielded many fine specimens of the larger forms, Rhinoceros, bovids 
and Stegodon .......... 524 

CXX. — (A) Flag and banner on the Yen-ching-kou temple. The banner reads 

"American Museum Representative" . . . . . -525 

(B) On top of the Dragon Bone ridge. Party returning to Yen-ching-kou 
after a three days' trip along the line of workings. The three rear 
coolies bear baskets with fossils . . . . . . -525 

(C) On the Dragon Bone ridge. A highly fossiliferous pit in working 
order. The children and women are engaged in gleaning fragments 
of bone overlooked or discarded by the workmen. Several specimens 

of the giant panda came from this pit . . . . . -525 

(D) Three tons of dragon bones. Piled in the comer of a room at Pai- 
shui-chih. These were the property of a wholesale drug merchant of 
Wanhsien and represent the output of many pits. They were proba- 
bly all eaten by Chinese patients diuing the following year or two 




(D) Women of the Tai tribe making offerings at a temple in Yuan Kiang 53 1 

CXXIII. — (A) Presbyterian Mission at Mosha on the Red River . . . -536 

(B) Expedition headquarters at the Presbyterian Mission of Yuan 
Kiang 536 

(C) Kapala. A mountain village of tribes people above the Mission at 
Yuan Kiang .......... 536 

(D) Village of Kapala. Potters at work with wheel. .... 536 

CXXI. — (A) A station on the railway line to Yunnanfu .... 

(B) "Victory tower," Yunnanfu ...... 

(C) Irrigation canal on the outskirts of Yunnanfu 

(D) Pack horses and their loads in a Chinese inn. Showing pack-train 
methods used in Yunnan ....... 

CXXII. — (A) Our caravan on the traU south of Yunnanfu .... 

(B) Chinese military escort on one stage of journey to Red River. 

(C) Suspension bridge over Red River above Yuan Kiang . 



CXXIV. — (A) One of the beautiful arched bridges common in northern Yunnan . 537 

(B) A Miao woman spinning hemp at the native village of Sapushan, 
near Wu Ting . . . . . . -537 

(C) Mr. Nelson at lunch in the Lung Kai temple. -537 

(D) Confluence of the Ma Kai River with the Yangtze -537 

CXXV. — (A) Expedition junk at entrance to the Ichang gorge .... 542 

(B) Junks sailing up the Yangtze at sunset . ..... 542 

(C) Expedition junk lying opposite the walled town of Kwei-chou . . 542 

(D) Crew of the Expedition junk and the landing plank . 542 

CXXVI. — (A) The walled town of Kwei-chou from a river steamer 543 

(B) Our boat at the upper entrance of Wushan gorge . . -543 

(C) Stepping stones across a tributary stream of the Yangtze . 543 

(D) Women washing clothes at Wanhsien landing .... 543 

CXXVn. — (A) Natural and Mantze caves — Wxishan gorge ..... 548 

(B) An occupied cave just above Wanhsien. ..... 548 

(C) Village of Kulowpei, below Ichang ...... 548 

In the foreground is a stretch of the shingle beach which yielded 

(D) Samples of the artifacts from the shingle beach shown in (C) . . 548 

CXXVIII. — (A) Cave and coffin cleft in the Wind Box gorge 549 

The trackers' path is seen cut out of the cliff. 

(B) Group of artificial caves at Fairy Bridge below Ichang . . 549 

(C) Fortress at base of sandstone clifT near Yen-ching-kou . . . 549 

(D) Neolithic site at Tai-chih 549 

The artifacts were found in the cultivated field. 








The original conception of the Central Asiatic Expeditions, of which this 
volxime is the narrative, took shape in my mind as early as 191 2. Since 1908 
I had been engaged in the study of Cetaceans, principally along the coasts of 
Asia, but I had decided to abandon that work for the broader field of scientific 
land exploration. 


I had before me the brilliant predictions of Professor Henry Fairfield 
Osbom,' made in 1900, that Asia would prove to have been a great dispersal 

Figure i. — Professor Osbom's Prophetic World Map of 1899-1900, Science, page 567, April 13, 1900. 

' Osbom, Henry Fairfield. 1900. "The Geological and Faunal Relations of Europe and America Dur- 
ing the Tertiary Period and the Theory of the Successive Invasions of an African Fauna," Science, N. S., 
XI, No. 276, April 13, pp. 561-574- 



center for northern terrestrial mammalian life; also my interest as a mam- 
malogist was stimulated by the extensive collections of new and little-known 
mammals being brought from China by the Duke of Bedford's Expeditions 
under the auspices of the British Museum (Natural History). 

In 1 91 5 I presented a plan to President Osbom for a series of expeditions 
which should extend over a period of ten years, and bring to the American 
Musevun of Natural History extensive zoological collections from Asia. 

As always, I was assured of President Osbom's enthusiastic support, and 
in 1916-1917 the First Asiatic Expedition, under my leadership, explored the 
little-known province of Yunnan, in southwest China, and the borders of Tibet 
and Burma. This Expedition not only brought large zoological collections to 
the American Museum but served to crystallize in my mind plans for work on 
a much broader scale. These collections have been described by Andrews,^ 
Bangs,^ Allen' and others. 

Having penetrated to that remote region after much difficulty and expense 
and no little danger, I found myself in the presence of the most interesting prob- 
lems in every branch of natural history; yet, because of lack of time and of 
highly specialized training in many sciences, I was unable to take advantage 
of the opportimities so lavishly offered. Moreover, in all my work as a zoolo- 
gist I had felt the lack of expert knowledge in other branches of science. Often 
puzzling faunistic problems presented themselves, which could have been 
solved if I had been a trained botanist. 

At the end of the First Asiatic Expedition, therefore, I had determined 
not to miss opportunities again by field researches in a single science, when 
such infinitely greater results could be obtained by proper organization of an 
expedition. Also, pur exploration along the edge of the great central Asian 
plateau had fired my imagination and enthusiasm to launch an attack upon 
that land of mystery; to discover whether or not it was the mother of north- 
em mammalian life. 

Service in the World War prevented the immediate execution of the plan, 
but at the end of the war I was in China with the added experience of having 
made several trips across Mongolia. The coimtry impressed me greatly. 
I was convinced that it was the place in which to start the scientific attack 
upon central Asia. With the view of obtaining a more intimate knowledge of 
that region, I cabled President Osborn asking permission to carry on zoological 
work there during the siommer of 1919. Thus the Second Asiatic Expedition 
came into existence. I spent six months in the desert, grasslands and forests of 

•Andrews, Roy Chapman, and Andrews, Yvette Borup. 1918. "Camps and Trails in China." D. 
Appleton & Co., New York. 

^ Bangs, Outram. 1 921. "The Birds of the American Museum of Natural History's Asiatic Zoological 
Expedition of 1916-1917," Bull. Amer. Mus. Nat. Hist., XLIV, pp. 575-612. 

' Allen, Glover M. 1923. "New Chinese Bats," Amer. Mus. Novilates, No. 85, pp. 1-8. 










2 *■ 

C o 

^ > 


northern Mongolia and obtained a large collection of mammals representing 
three faunal areas. A matter of much greater importance, however, was what 
I learned of the coiintry itself. 

At that time a few motor cars were running irregiilarly between Kalgan, 
China, and Urga, the capital of Mongolia. The old caravan route was difficult 
in places and the Ford motor cars which were first vised had proved unable to 
meet the severe demands made upon them. Accidents were frequent and 
many people had been killed and injured, but Mr. Charles Coltman (who was 
murdered by Chinese soldiers in 1923) had successfully taken several Dodge 
Brothers cars to Urga. The fenceless rolling grasslands and the gravel desert 
made it possible to run off the trail at will and gave me the idea of conducting 
explorations by automobile. In Tibet and Chinese Turkestan, camels, mules 
or yaks must be relied upon, but I was certain that motor cars coidd be used 
successfully even in the western Gobi. 

I retiimed to New York early in 1920 with the basic plans of the Expedi- 
tion clearly in mind. The main problem was to be a study of the geologic his- 
tory of central Asia; to find whether it had been the nursery of many of the 
dominant groups of animals, including the human race; and to reconstruct 
its past climate, vegetation and general physical conditions, particularly in 
relation to the evolution of man. We were to bring to bear upon our problem 
every branch of science which could possibly assist in its solution. Moreover, 
these sciences must be represented by men of the highest scientific ability. 
We must take the men into the field together, so that each would have the 
advantage of assistance from the others ; correlated work was to be the basis of 
the scientific organization. Motor cars were to be used in conducting the 
exploration while a caravan of camels transported the supplies to specified 
locations. This procedure would give us the great advantage of speed over 
previous explorers. Headquarters were to be established in Peking, where the 
Expedition was to be prepared for actual field operations and where the com- 
plicated diplomatic and administrative work could be carried on. 


Such was the 1920 outhne of the Expedition. After careful consideration 
by the President and Trustees of the American Museum, it met with their 
enthusiastic approval, provided I would undertake to obtain the financial 
support outside of the Museum, as well as to organize and direct the Expedi- 
tion. Thus, raising the money rested upon my shotilders. Two hundred and 
fifty thousand dollars and five years was what I considered necessary in money 
and time to conduct the work effectively. Later the time was extended to ten 
years and the finances to six hundred thousand dollars. 

The American Musetun pledged itself for five thousand dollars a year 


and the services of such members of its regtilar staff as should be detailed to the 
work. The American Asiatic Association and its official organ, Asia magazine, 
contributed a total of thirty thousand dollars; and after 1924 the Field Mu- 
seum of Natural History, Chicago, gave five thousand dollars a year, with the 
understanding that they should have the first series of duplicates of our col- 
lections. The remaining fvmds all came from private individuals whom I was 
able to interest in the proposed Expedition. At present, thirty-six states and 
the District of Columbia are represented among the six himdred and fifty-five 
subscribers; four contributions are from England, one of which is from the 
Central Asian Society of London; two from China; two from Honolulu and 
one from each of the following countries: Porto Rico, Switzerland and Nor- 
way. A few of the subscriptions are large, but most of them are for small 
amounts. I tried to make it clear to ever>-one that we were playing an "off 
chance" in the scientific race; that the dividends would be large if we won but 
that by the same token the result might be entirely negative. 


It was tinnecessary to urge the importance of the project upon any scien- 
tist, for the fossil history of central Asia was completely unknown. The British 
had made some discoveries of interest in India. In China fossil teeth and bones 
had long been sold for medicine in the drug stores, and a few of these had been 

Figure 2. — Map of Mongolia superimposed upon map of United States to show comparative size. 


studied and described by Owen, 1870; Lydekker, 1885 and 1896; Gaudry, 
1872; Koken, 1885, and Schlosser, 1903. By far the most important work was 
that of Schlosser, but all his material was purchased in drug shops and con- 
sisted mostly of isolated teeth without data. Some interesting and important 
finds had been made in Java, Persia and Asiatic Russia, but these all came 
from the outskirts of the great continent. Literally the only fossil known from 
the central Asian plateau was a single "rhinoceros" tooth discovered by the 
Russian explorer Obruchev^ in Mongolia during 1 894-1 896. 

As a matter of fact, when it was announced that we were to explore the 
Gobi Desert for fossils, where attempts in that direction already had been 
made, our project was ridiculed by some other scientists, particvilarly in Peking. 
They pointed out that we wovild find only a waste of sand and gravel and that 
we might as well search for fossils in the Pacific Ocean. Our attention was 
called to the fact that none of the other geologists who had crossed Mongolia 
had discovered bones, with the single exception of Obruchev. Also I was told 
it was little short of criminal to waste the time of such eminent geologists as 
Berkey and Morris in a country where the geology "was all obscured by sand." 

However, I felt that I knew Mongolia better than ovir critics; that in many 
instances the work of previous explorers had been too much political and too 
little scientific ; that they had not been able to use the modem methods which 
we intended to inaugiurate — in short, that what had been done in the past 
afforded no criterion as to what Mongolia might yield to our scientists. 


Although Mongolia had been crossed and recrossed by some excellent 
explorers, mostly Russian, no part of the country had been studied by the 
exact methods of modem science. Foiir primary reasons were responsible. 

First, Mongolia is isolated in the heart of a continent, and, until recently, 
a considerable journey was required even to reach its borders. (See Relief 
Map of Mongolia at end of volume.) Second, the distances are great and 
transportation slow. Mongolia is roughly two thousand mUes in length 
from east to west and twelve hundred miles in breadth from north to 
south. In all this vast area there is not a single mile of railroad. Trans- 
port except in the desert is by camels, ponies and oxcarts. In the Gobi, 
which extends from west to east through the heart of Mongolia, only camels 
can be used throughout the year. A camel caravan moves at the rate of two 
and one half miles an hour and averages about ten miles a day. Third, the 
climate is very severe. During the winter the temperature drops to forty or 
fifty degrees below zero Fahrenheit, and the plateau is swept by bitter winds 

' Suess, Ed. 1899. "Ueberreste von Rhinoceros sp. aus der ostlichen Mongolei," Mit Anmerkungen 
von W. Obrutschew, Verhandl. KK. Russ. Mineral. Gesell. St. Petersburg, Zweite Serie, Band XXXVI, No. 2. 


from the Arctic Ocean. Most types of scientific investigation are impossible 
at that time, and bare existence demands the strongest constitution. Effec- 
tive scientific work can be conducted only from April to October. Fourth, in 
the Gobi, which occupies a large part of Mongolia, food and water are scarce, 
and the region is so inhospitable that there are very few inhabitants. 

Motor transport as a solution 

I felt certain that all the physical difficulties could be solved by some 
means of rapid transportation and that without it an expedition such as we 
had in mind could not be carried on successfully. I believed that the auto- 
mobile was the answer to the problem. With motors we could go into the 
desert as soon as the heavy snows had disappeared, penetrate to the farthest 
reaches of Mongolia, and return before continued cold and snow set in. Thus 
we should escape the six months of devastating winter when such scientific 
researches as we had in mind wovild be impossible. From previous experience, 
I believed that in cars we could travel about one hundred miles in a day ; that 
is, ten times as far as a camel caravan. Thus we should be able to do approxi- 
mately ten years of work in one season. As resvilts proved, we did maintain 
just about that ratio. Moreover, our investigations were made much more 
effective than they could have been otherwise, for the geologists and palaeon- 
tologists were able to leave the main fleet at any time to examine outcrops or 
exposures which were fifteen or twenty miles away. With camels this would 
have been slow and laborious. 

The great aridity and scarce water supply did not greatly trouble us. 
Often we crossed four or five hundred miles of gravel desert where it was well- 
nigh impossible even for camels to exist; for with our cars this meant only a 
few days of travel. The weUs on the main caravan trails are usually ten to 
twenty miles apart, but we could carry sufficient water for several hundred 
miles. Thus it may be seen how easily rapid transport solved the physical 
difficulties of exploration in ^Mongolia. (See Route Map of Mongolia at end 
of volume.) 


Although the physical difficulties of maintaining a large expedition in such 
a remote region as the Gobi were great, they were insignificant in comparison 
with the political obstacles which had to be overcome before we could begin 
work. In the winter of 1 920-1 921, Baron Ungem-Stemberg assisted the 
Mongols in driving out the Chinese. Baron Ungem was defeated by the 
Bolshevik troops a few months later and the Soviet Government sent soldiers 
to himt down the White Russians in Mongolia. The Soviets almost imme- 
diately organized a political party of Red Mongols who established a govern- 


ment which became more and more a tool of the Soviet. In 1922 the Soviet 
influence was already strong; by 1923 it was steadily growing and at the time 
of our Expedition in 1925 the government of Mongolia was entirely in their 
hands. Although a Mongol Premier and Cabinet were nominally maintained, 
Buriat and Russian "advisers" controlled the affairs of the country. 

It seemed to be impossible for the authorities to understand that we were 
in Mongolia for scientific exploration and nothing else. We were suspected of 
political intrigue for the Chinese; of being a commercial expedition in search 
of gold or oil, masquerading tinder the cloak of science; of being spies sent by 
the American Government to prepare the way for a conquest of Mongolia! 
Everything was believed of us except that we were there for scientific explora- 
tion, the purpose we had stated. 

It required weeks of work to allay their suspicions so that we might pro- 
ceed, and many days to ftilfill the complicated requirements of passports and 
other impedimenta. Each year the conditions were increasingly difficult. 
Had it not been for the assistance of Mr. F. A. Larsen, Mr. T. Badmajapoff 
and of many Mongol friends who had known me in former years, the Expedition 
never could have gone into the field. 


I had great faith in the method of correlated work. It never had been 
attempted on the scale on which we intended to operate, and it resvdted in 
practice even better than it gave promise of doing in theory. As we sat in the 
mess tent at night discussing the day's work, it was most interesting to see 
how puzzling situations in geology would be clarified by the palaeontologist- 
how the topographer brought out important features which gave the key to 
physiographic difficulties; and how the palseontologist would be assisted by 
the palaeobotanist or geologist in solving stratigraphic problems. It is doubt- 
ful if our archaeologist ever could have dated the cultiire and unraveled the 
story of the Dune Dwellers without the assistance of the other scientists of 
the Expedition. Moreover, I believe that this type represents the exploration 
of the future. To-day there remain but a few small areas on the world's map 
unmarked by explorers' trails, or which have not been seen from an aeroplane. 
Human courage and endiorance have conquered the poles; the secrets of the 
tropical jungles have been revealed. The highest mountains of the earth have 
heard the voice of man. This does not mean, however, that the men of the 
future have no new worlds to vanquish; it means only that the explorer must 
change his methods. 

Intensive scientific investigation 

We stand on the threshold of a new era of scientific exploration which is 


just as romantic, just as alluring and just as adventvirous as that of Peary 
and Amundsen, of Stanley and Hedin. In almost every country of the earth 
there lie vast regions which are potentially unknown. Some of them are 
mapped poorly if at all, and many hold iindreamed-of treasures in the realm 
of science. To study these little-known areas, to reveal the history of their 
making and interpret that history to the world to-day, to learn what they can 
give in education, culture and for himian welfare — that is the exploration of 
the future ! 

That the growing importance of intensive scientific exploration of a given 
region is being recognized by the geographical societies is shown by the intro- 
ductory remarks of the late Dr. D. G. Hogarth, President of the Royal Geo- 
graphical Society, upon the occasion of my address to that distinguished body 
in London on November 8, 1926. The President spoke in part as follows: 

"He is going to tell us to-night about . . . the organization of almost a 
new form of exploration — the realization of what I ventured in my presidential 
address a few months ago to hope would be the work in the future of this 
Society — that is, intensive study of a particiilar area. . . . This is, therefore, 
a very notable moment in the history of the Society; it is about to hear what 
the exploration of the future is to be. The great journeys of exploration are 
probably done in almost all the world ; the great pioneer lines have been thrown 
across the continents. What we have to do now is to fill up great gaps between 
those single lines. "^ 

Reconnaissance and specialization 

It would have been ruinous to bring a group of highly trained men from 
America to Mongolia at great expense unless I knew definitely that they could 
do effective work. The region we were to explore was virtually vmknown from 
the standpoint of oiu- investigations. Therefore, it was obvious that the first 
season must be strictly a reconnaissance. We must make a general survey of 
the country without attempting exhaustive work in any one place. The next 
year then could be one of intensive research, and the amount of time to be 
devoted to each locality carefvdly planned in advance. Such intensive work 
would require a different staff from that for rapid reconnaissance. 

As an example, in the summer of 1922, Dr. Walter Granger was the only 
palaeontologist. On the next expedition he was reinforced by three other 
expert fossil collectors and two trained Chinese assistants, because such rich 
fossil deposits had been discovered that we needed a large staff for their explora- 
tion. We planned in advance the number of days or weeks that could be 

' Hogarth, D. G. 1927. Introductory remarks of the President of The Royal Geographical Society to 
Roy Chapman Andrews' paper: "Explorations in Mongolia: A Review of the Central Asiatic Expeditions of 
The American Museum of Natural History," Geog. Jour., LXIX, No. I, Jan., pp. 1-23. 


devoted to each locality and with one exception maintained the schedule 
throughout the summer. This method of reconnaissance, followed by intensive 
work, is the only effective way to avoid loss of time and effort. 

In 1922 the Expedition comprised the following sciences: palaeontology, 
geology, topography, photography and zoology, the latter including mam- 
malogy, herpetology and ichthyology. Mr. Clifford H. Pope, who devoted 
his attention to the two last named sciences, did not go to Mongolia but car- 
ried on his work entirely in China proper. 

In 1923 we concentrated on paleontology, dropped the photographer and 
one geologist and did httle topography. In 1925, another reconnaissance year, 
we added one more motor expert, three topographers, a palaeobotanist, an 
archaeologist, and brought back the chief geologist. The 1925 Expedition totaled 
forty men. The foreign staff included twelve Americans and two British; and 
in the native personnel there were thirteen Chinese and thirteen Mongols. 
In 1928 the Expedition consisted of twenty- two men, ten of whom were 

Field units 

With so many sciences represented, it was improbable that all the men 
could find work in the same place at the same time. Therefore, the Expedition 
was divided into several units, each with its own driver, Mongol interpreter, 
cook and camp gear. Any unit could leave the base camp and maintain itself 
independently for several weeks if necessary. As a matter of fact, the Expedi- 
tion was almost continually divided except when we made long marches to 
new localities; frequently there were four camps from twenty to a hundred 
miles apart. 

Laboratory arrangements 

The expense and time involved in preparing the fossil material for exhibi- 
tion or study was a great problem at the American Museum. The Chinese 
are natioraUy so skUlftd with their hands that we decided in the winter of 1925- 
1926 to have the work done in Peking. Consequently, a well-equipped labora- 
tory was established at the headquarters under the direction of Mr. George 
Olsen. He trained o\ir taxidermists and other assistants, and the final prep- 
aration of most of the 1925 fossil collection was done in Peking at about one 
fifth what it would have cost to do the work in New York. The 1928 collection 
was likewise prepared in Peking, under the direction of Mr. Albert Thomson. 


The organization and equipment of the Expedition had to be carried out 
simultaneously with the efforts to finance it. Of course, selecting the staff was 


the most important single task. The general fitness of a man for the work, 
as well as his scientific training, needed to be carefully considered, for person- 
ality, character and the abihty to get on with other men are, perhaps, the most 
important factors in determining the success or failure of such an expedition. 

I may say that our staff could not have been more happily chosen, for dur- 
ing the five seasons in MongoUa there was not a single quarrel. We worked 
together with a harmony, mutual confidence and lack of personal jealousy 
which are rare in any organization. 

The permanent native staff developed more slowly but not less success- 
fully. As a basis, I had several Chinese assistants who had worked with me on 
former expeditions. These were added to as the 1 92 1 preliminary trips in China 
proper brought out men especially adapted for certain types of work. Even- 
tually two of our young Chinese, Kan Chuen-pao and Liu Hsi-ku, showed 
such unusual ability in fossil collecting that we sent them to New York in 1923- 
1924 for a year's training in the palseontological laboratories of the American 
Musevun of Natural History. The experiment was entirely successful and 
they retvimed to us in China as expert field and laboratory men. Three other 
Chinese, Kang Fu-min, Chih Sho-lun and Chang Yu-lu, I personally trained 
for zoological work. They became highly efficient in collecting and preparing 
mammals and birds and later learned the technique of fossil hunting and 
laboratory preparation. The three cooks and two mess boys were not less 
efficient in the difficult task of ministering to the wants of the foreign staff. 
The head cook, Whey, died in 1926, while on a field trip in Yunnan with 
Messrs. Granger and Nelson. To the end he was loyal to his duty and to the 

The Mongol caravan men and the three interpreters were old employees 
of Mr. F. A. Larsen of Kalgan, who engaged them for us. They became 
devoted members of the Expedition and gathered their families into a single 
large village, of which Merin, the caravan leader from 1922 to 1927, was chief. 
This splendid native staff, both Chinese and Mongol, was a factor in the 
success of the Expedition. 


When it was announced that we were to attempt a scientific exploration of 
the Gobi Desert with a fleet of motor cars, even those men who had driven from 
Kalgan to Urga said that we were little less than fools. They advanced a dozen 
reasons why such an expedition covdd not succeed, but in my opinion every 
one of them was answerable by preparation and organization. If we took with 
us a complete assortment of spare parts and expert mechanics who could 
repair breakage, I felt that we had reasonable chances of success. 

For travel in the Gobi, a car must be light, have a high clearance, great 



O O 

* •§ 


P o 

5 K 

< M 

a .£ 

H « 

< W 



durability, a flexible chassis and not less than a twenty-eight horsepower 
engine. The Dodge Brothers cars fulfilled these specifications to the letter. 
We also piu-chased two one-ton Fulton trucks, but, although these gave excel- 
lent service, we eventually sold them in Kalgan because they were too heavy. 
In certain parts of the desert there is a hard-baked surface of coarse sand. 
This supported the light Dodge Brothers cars, but the heavier Fultons were 
continually breaking through the crust. On the 1922-1923 expeditions we 
used three Dodge Brothers cars and two Fulton trucks. In 1925 the fleet 
was increased to seven, by the addition of two Dodge Brothers cars; in 1928 
the fleet consisted of eight Dodge Brothers motors. 

The first cars were all ordinary stock cars; for the 1925 Expedition cer- 
tain changes were made as a result of our two years' experience. The autos as 
then used we consider ideal for Gobi travel. All except one had an open 
"express" body with eight-inch sides of heavy screen wire. The springs, 
both front and rear, were heavier than on the ordinary commercial car. In 
addition, on each rear spring, inside, we had iron btimpers lined with pieces 
of heavy outer tire-casing. Thus, the body of the car when heavily loaded 
rested on this additional support, making it almost impossible to break a spring 
when traveling over rough ground. Leather snubbers were also affixed to 
prevent too great a rebound. The gasoline tanks were increased to contain 
twenty-one gallons, and four strong hooks were bolted onto the chassis mem- 
ber, front and rear, to facilitate pulling out of mud and sand; also two com- 
plete spare wheels were carried on each car, one on either side of the driver's 
seat. We had one car with an ordinary touring five-passenger body. This 
being the lightest, I always used it to drive several miles in front of the fleet 
to scout the road, and for advance explorations. 

We foxmd that the 33 x 4.5 Royal Cord tire, made by the United States 
Rubber Company, gave the best service. BaUoon-type tires were not practi- 
cable, as we discovered by one season's trial in 1925. Although they held 
the car up better in sand, they increased the gasoline consumption to such an 
extent that this alone would rtde them out. In addition, they were too 
easily cut by stones and roots. We carried hundreds of nuts and bolts, almost 
every conceivable spare part, and the very best tools; our motor experts were 
highly trained men; thus, short of actual wrecking of the chassis or engine, 
we were prepared for any emergencies. 

The success of the transport is shown by the fact that we used the same 
fleet of cars for the first two expeditions; that we traveled more than six 
thousand miles over a virtually unknown country where there are no made 
roads; that the Dodge Brothers cars did as much as ten thousand miles; that 
when we returned all the motors were sold in Kalgan, as they stood, with no 
repairs, within three days, for m.ore than they cost when new in America; 


moreover, that the same fleet continued to do service on the Kalgan-Urga 
trail, in the hands of a Chinese company. The record speaks for itself. 

Oxir explorations had the unexpected result of opening western Mongolia 
to motor transport. Immediately after our return from the first Expedition, 
fur dealers came to ask how they coiold reach certain far-distant trading sta- 
tions. We told them where to send gasoline and how they could go; now fur 
and wool buyers who cross the desert use dozens of cars, where only camels 
had traveled untU we came. 

Gasoline cases 

Packing the gasoline was a great problem, and even yet it has not been 
solved to our satisfaction. Four thousand gallons were sent out for the 1925 
Expedition. Two five-gallon tins fitted into a wooden case which weighed 
sixty-five poimds; each camel carried six cases. The leakage was great. 
Straw was wedged between, as well as around the sides of the tins, but even 
that did not prevent friction. Wherever the straw wore through and allowed 
the tins to rub against each other, breaks were almost certain to develop. 
The great difference in temperature between the fierce day-heat of July and 
August and the cool nights produced a constant expansion and contraction 
of the tins which was ruinous. Frequently when we opened a case one or both 
of the tins would be empty or only partially filled. Of course, at every meet- 
ing with the caravan we inspected the gasohne carefully and took the lightest 
cases into the cars to be first used. The leakage on the 1922 Expedition was 
nearly fifty per cent; in 1923 it was reduced to thirty and in 1925 to twenty- 
five per cent. 

The difficulty might be overcome to a certain extent by using extra hea\^ 
tins, but this entails a large expense in having both tins and wooden cases 
specially made. We considered the advisability of taking the cylindrical 
steel drums which are sometimes used for shipping gasoline to the Orient. 
The difficvilty, however, of packing cylinders on camels, the initial expense of 
purchase, the extra weight and the fact that they must be carried back decided 
us against their use; again, the wooden cases were very important as fossil 
containers and if they were not available, boards must be carried with which 
to make boxes. It was cheaper to send out enough extra gasoline to provide 
for the leakage. 


We have been referred to as a "de luxe" expedition because we had the 
maximiim of camp comforts. I may say, however, that there was not one 
item of unimportant equipment. In the field we were a group of men who 
worked at high tension for nearly six months. An army cannot fight imless 


it is well fed; neither could oirr men have accomplished what I expected of 
them unless they were kept physically fit. 

I do not believe in hardships, if they can be avoided, for they lessen effec- 
tiveness; they are a great nuisance. Eat well, dress well and sleep well is a 
pretty good rule for everyday use. Don't court hardships. Then you can 
work hard and steadily, and, if a bit of "hardship" does happen in the coiu-se of 
things, you are ready to take it in yoiu- stride and laugh while it continues. 
With us it simply meant sending out a few more camels to carry the extra load 
of supplies, food and eqioipment which made just the difference between com- 
fort and discomfort. 

Neither do I believe in adventures. Most of them can be eliminated by 
foresight and organization. My friend Stefansson, the Arctic explorer, has a 
motto which I am very fond of quoting because it expresses a great deal in a 
single sentence. He says "Adventures are a mark of incompetence." If the 
explorer has a clear-cut problem to solve, and an honest desire to contribute 
something of worth to the world's knowledge, he wiU prepare against adven- 
tvures. It will disappoint the newspapers, but facilitate his work. How infi- 
nitely more creditable it is to eliminate difficiolties through foresight and prepa- 
ration before they are encountered, than to suffer heroically and leave the work 
half done ! 


Most explorers find that in every country the natives have developed the 
best type of dwelling and of clothing for their particular conditions. This is 
true of Mongolia ; the Mongols are nomads who are constantly moving as they 
foUow their flocks or the dictates of their restless spirits. A permanent dwell- 
ing would be of little use, as the grazing may be good at a certain place one 
year but poor the next. Wind and cold are the most serious weather conditions 
to be met, for there is comparatively little rain in the grasslands or desert. 
The Mongol tent and the felt-covered yurt are ideal. 

The tent is usually made of two thicknesses of blue cotton cloth, light in 
weight but not very waterproof. The sides sweep down to the ground from the 
ridge-pole in long curves which present sloping stu"faces to the wind at every 
angle. Thus if the tent is firmly pegged it will resist a heavy wind. The 
cloth may rip but usually the tent will remain standing. It can be erected in 
a gale when it would be impossible to pitch a wall-tent. A man can put up a 
small tent alone. First one side is pegged down and the ridge and poles put in 
while it Ues upon the grotmd ; then with a rope the tent is pulled upright. It 
will stand in position while the other side is fastened. If a fire is built a little 
beyond the center and one edge of the tent elevated to create a draft, the 
smoke will be carried out through the door. During the svimmer all the edges 


can be raised, converting the tent into a pavilion. For a windy, open country 
where there is little rain, this is an ideal tent. 

The yiirt is an extraordinarily practical and comfortable dwelling. It is 
circixlar with a cone-shaped roof and thus has no flat surfaces to resist the 
wind. It can be erected in thirty minutes; in the same time it can be taken 
apart and packed upon a camel. The framework consists of wood lattice 
sections which can be folded like an umbrella ; the felt covering is so excellent 
a non-conductor that when a fire is Ughted in a sheet-iron stove or in the open 
brazier the yurt is warm even though the temperature drops to forty degrees 
below zero. In hot weather the felt sides may be roiled up or removed alto- 
gether and the yurt is delightfully cool. 


Our summer clothing was the ordinary'' flannel shirt and khaki breeches, 
but the Mongol coat and trousers of sheepskin, wool inside, can hardly be im- 
proved upon for winter. We used felt hats and did not find sun helmets 
essential. Sheepskin sleeping-bags are indispensable. No blankets can keep 
one warm. We had cot beds for the base camp but when off on reconnaissance 
trips we slept on the grovmd. Personally I prefer to do so always. Folding 
tables and chairs were a great comfort for the main camp. 


• I selected the food with a great deal of care in order to give as varied and 
healthful a diet as possible. The Mongols live almost entirely upon meat and 
animal products and raise no grain or vegetables. I brought dried tomatoes, 
carrots, spinach, beets and onions from New York and we took enough fresh 
potatoes to last a month. Dried milk and egg powder were easily carried and 
very welcome. Dried lemon powder was one of our greatest comforts, and made 
a delicious and healthful drink. We carried no spirits except a few bottles of 
whiskey for birthday celebrations. This varied diet, the dry clear air, strong 
sunlight, sparse popvdation and consequent lack of germs, kept all our party 
in excellent health. On the expeditions, we had virtually no sickness among 
either the foreign or native staff. 


Very soon after the Expedition went to Mongolia, we discovered that my 
original paper plan for field work was sound in practice. The supporting 
caravan, carrying gasoline, food and other supplies, was dispatched several 
weeks in advance of the motor party. Its objective was a well-known 
place in the desert, six hvmdred miles from Kalgan, our starting point. The 
camels traveled slowly, leaving one or two depots of food and gasoline on their 





< u 

jA«CT»i:..ttit; : -, 







trail, and had some time to rest before the motor party arrived. Upon con- 
necting with the caravan, we transferred to it all equipment which was not in 
daily use and took in the cars enough food and gasoline to maintain us until 
oiu" next meeting with the camels. 

Instructing the caravan to follow rapidly, we proceeded in the motors to 
a suitable region for work and estabUshed a central camp. Investigations 
were carried on in a wide circle about this base. In the meantime, I made 
advance trips of several himdred miles with one car, exploring the country and 
estimating its value for future research. When all the scientists had com- 
pleted their studies at the central camp, the Expedition moved forward to the 
new region which had been selected and operations were continued. Usually 
the caravan was behind us, but at times it could be sent on ahead when our 
moves were short and we remained for a considerable time in one spot. The 
success of the motor transport depended, of course, upon maintaining touch 
with the caravan. 

At each meeting with our caravan we took supplies of gasoline and food 
and gave it the collections which had been made. The caravan boxes 
were especially constructed with sUding tops and the wooden gasoline cases 
were of convenient size for packing specimens. We used the supplies a little 
faster than collections were gathered so that there were always several extra 
camels. This was desirable as sometimes many camels died when it was neces- 
sary to send them across long stretches of particularly arid desert. 


The average altitude of the Mongolian plateau is about forty-five htmdred 
feet. The temperatiu-e diuing the winter drops to forty or fifty degrees below 
zero Fahrenheit, and terrific gales sweep down from the Arctic Ocean and the 
Siberian steppes. There is a good deal of snow, which is blown into great 
drifts and makes traveling diflficvilt even for camels. We could not do our work 
effectively tmtil the snow had disappeared; therefore, our season was limited 
to the summer months between April first and October first. About the mid- 
dle of April there is a fortnight of delightfiol weather with warm sun and little 
wind. We always took advantage of that period to get well into our field of 
operations. The first week of May usually sees a return of cold and wind with 
occasional snowstorms; such weather persists tmtil the middle of Jtme when 
the real heat begins. 

In July and August the temperature dtiring the day may reach no degrees 
Fahrenheit in the shade and 145 degrees in the sun, but the air is so dry that 
the heat is not debilitating. The nights are always cool or even cold and it 
was seldom that our sheepskin sleeping-bags were not welcome. The sun 
almost always shines, and on the desert rain is so infrequent that it hardly 


need be reckoned with in preparations for an expedition. In the grasslands 
north and south of the Gobi there is considerable rain during late July and 
August. It usually comes in torrential downpours which are soon ended. 


The heavy mortality among our camels was due to the fact that we had 
to travel during the svimmer. The Mongolian camel is a cold-weather beast; 
he cannot stand the heat. He should feed all summer, storing up fat in his 
hiunps, and work during the winter. We had to reverse the travel season, 
with the result that during the winter the animals could not find sufficient 
food to put them in good condition for the next summer. Therefore, each 
spring it was necessary to buy thirty or forty new camels to replace those 
which had died or were too weak to endure a long trip. It entailed heavy 
expense but was tmavoidable, for hiring a caravan was impossible. The 
price per camel varied with the political situation and the consequent trade. 
When conditions were fairly peaceful and much merchandise was being 
transported, we paid as high as seventy dollars. United States currency. At 
other times we purchased camels at forty dollars each, 


The close association which the Expedition formed with the Geological 
Survey of China and the Geological Society of China in Peking contributed 
much, in the first years, to our work. The Geological Survey had been conduct- 
ing palasontological researches in China proper for four years before we arrived. 
These were largely tinder the direction of Dr. J. G. Andersson, and were made 
possible by fvmds from Sweden, provided by a research committee headed by 
H. R. H. the Crown Prince of Sweden. In order not to duphcate their work, 
and create unhealthy competition, we agreed to confine our explorations to 
Mongolia or to such parts of China as the Survey designated. This arrange- 
ment itself worked admirably, and we had, until 1928, the most cordial coopera- 
tion from the Geological Survey, the Geological Society and the government 

The first year the Geological Survey offered us a fossil locality on the 
Yangtze River at the village of Yen-ching-kou, not far from Wanhsien. Our 
palseontologists worked there for three winters and would have continued for 
another season had not war and bandits prevented. During the winters I 
usually returned to America to obtain additional fimds for the Expeditions and 
to stimulate public interest in our work by extensive lecture trips. 



We coiild not consider work in Mongolia during the first summer, because 
it was necessary to make diplomatic arrangements and get the complicated 
machinery of such a large expedition under way. Therefore, I sailed several 
months in advance of the main party and arrived in Peking April 14, 1921. 

I was fortimate in finding an ideal house for the headquarters of the 
Expedition. Its former tenant, my old friend, the late Dr. G. E. Morrison, 
was one of the best -known Britishers who ever lived in north China. His mag- 
nificent Ubrary, his brilliant writings for the London Times, his fascinating 
personality and his interest in science and exploration made his house a Mecca 
for travelers of every nationality. 

Like all Chinese houses, the compound, which included one acre, was sur- 
rovmded by high walls. Soon it became a small city in itself devoted to the 
multiple interests of the Expedition. There were the living quarters of my 
own family, garages for eight cars, stables, a house for the storage of equip- 
ment, an office, laboratories and a complete motion-picture studio. 


It was not vrntil September that the house was finally prepared for occu- 
pancy, and in the meantime (Jime 28) Walter Granger, Chief Palaeontologist, 
and Clifford H. Pope, Herpetologist, had reached Peking. As I have remarked 
in Chapter I, the Geological Siirvey of China had invited us to investigate a 
reported fossil locality near Wanhsien, on the Yangtze River. Granger 
departed thence in August and remained there all winter. Since I never visited 
the region I shall leave to him the narrative of his own work. I may say, 
however, that he carried on under the most difficult and dangerous circum- 
stances. The fact that he was able to do it so successfully indicates that he 
should have honors for courage and diplomacy, as well as for paleeontology. 

Since Mr. Pope was unfamiliar with the Chinese language and the meth- 
ods of collecting fish and reptiles in China, I took him with me on a short 



expedition to the Eastern Tombs (Tung Ling) region, eighty miles from Peking. 
This also gave me an opportimity to train several Chinese assistants in zoologi- 
cal collecting. My plan was to have Mr. Pope make a siirvey of the herpetology 
and ichthyology in every province of China proper, because MongoUa is so 
cold and dry that its reptilian favma is exceedingly limited and I could do the 
necessary collecting there. The results of Pope's careful and enthusiastic 
labor already have produced, by far, the largest and most complete collection 
of fish, reptiles and batrachians that has ever been made in China. 

After returning from Tung Ling in August, 1921, Pope went to Anhwei 
Province and spent the winter of 1921-1922 in the region of the interesting 
Tungting Lake, Hunan. He made an expedition to Shansi and the border of 
the Ordos Desert in the svmimer of 1922. He spent the year 1923 on the little 
known island of Hainan, southwest of Canton. Later he collected for many 
months in Fukien Province. At times he conducted his work under the most 
dangerous circumstances. In 1922, in Shansi, he was in a city that was cap- 
tured by bandits, yet, by his tact and courage, he not only saved his life and 
collections but continued his work. On the island of Hainan it was highly 
dangerous to go beyond narrowly circumscribed limits, because the region 
swarmed with brigands; yet he remained a year and brought out a superb col- 
lection. He learned the difficult Chinese language so well, and has such a sym- 
pathetic and thorough understanding of the people, that I consider him one of 
the best equipped field men for that country whom I have ever known. 

In September, 1921, I made a month's expedition to the Tsinling Moun- 
tains of Shensi Province, to collect a group of the rare takin, Btidorcas bedfordi, 
for the Hall of Asiatic Life in the American Museimi of Natural History. 
In November, of the same year, I went to north Shansi for bighorn sheep, 
Oins animon dannni, and stag, Cen<us canadensis kansuensis. 


The remainder of the winter was spent in making the final preparations 
for the first Expedition into Mongolia. Everything that could possibly go by 
the caravan was packed in boxes in order to reduce the load for our cars. The 
camel boxes were thirty-four inches long by seventeen inches wide and twenty- 
two inches deep, and had sliding tops. This we have found by experience to 
be the best size for packing fossils, for a box must not exceed two hundred 
pounds in weight. The boxes went out filled with food and supplies but 
returned carrying specimens. Every package had to be weighed, roped and 
invoiced, so that we knew in which box each article had been placed. When 
all was ready, the gasoline and equipment were sent by rail to Kalgan, one 
hundred and fifteen miles from Peking. 

At Kalgan, my old friend, Mr. F. A. Larsen, then manager of Anderson, 



FXtr.rpiTiiiv; \T J Kisr, Hfii 

., CHINA. 
i| Darters. 


Meyer and Company, placed himself and his great compotmd at our disposal. 
Through him I purchased camels, selected the Mongol caravan men and 
attended to the countless other details incident to the proper equipment of a 
caravan for such a long expedition into the desert. 

On March 21, 1922, our seventy-five camels left Kalgan. The instructions 
were to proceed along the Kalgan-Urga trail, leave a supply of gasoline at 
Iren Dabasu (Erhlien), and await us at Tuerin, a large monastery one hundred 
and fifty miles southeast of Urga. They were to arrive at Tuerin on, or before, 
April 25 ; this gave them five weeks in which to travel five hundred and eighteen 
miles. Since the camels were in good condition, we felt that they could maintain 
this average of fourteen miles a day. 

After the caravan had started, I returned to Peking to superintend the 
final preparations, arrange for shipment of the five cars to Kalgan, and attend 
to the multiple final details. The foreign staff had assembled in Peking by 
April I, 1922. Granger returned from Szechwan Province after a successful 
winter at the fossil pits of Yen-ching-kou. Berkey, Colgate and Shackelford 
arrived from America, and Morris came up from Tientsin where he had been 
teaching in the university. 

The headquarters seethed with activity. Every man was occupied with 
his own individual preparations for the long summer in the desert. The 
courtyard in front of the main laboratory was strewn with skins, boxes and 
equipment, which were being packed for shipment to New York or to go with 
us to Mongolia. Colgate had the front court filled with cars, and all day the 
whir of motors being tested and the ring of hammers made it seem like an 
open-air garage. As if to bid us Godspeed, the lilacs and flowering trees in 
the courtyards, in bloom almost a week earlier than in any other part of the 
city, transformed the compound into a veritable paradise. 

A farewell dinner was given us by Mr. Albert B. Ruddock, First Secretary 
of the American Legation, at which Mr. C. S. Liu, then Director-General of 
Chinese Railroads, became so much interested in our plans that he offered to 
send the motors and equipment to Kalgan free of charge and give us two pri- 
vate cars for the staff. His courtesy was doubly appreciated, because war 
clouds were gathering thickly in north China skies and continual troop move- 
ments made railroad transport most imcertain. There seemed to be little 
doubt that the expected clash between Chang Tso-lin and Wu Pei-fu would 
take place within a few weeks, as indeed it did. 


Kalgan is a Chinese city with a strong Mongol influence, and, as it is only 
ten miles from the foot of the passes which give entrance to the MongoUan 
plateau, it was an ideal advance base for the Expedition. The altitude at the 


railway station is twenty-four hundred and thirty-seven feet above sea-level. 
Kalgan is the most important gate of the Great Wall and for centuries has been 
a focal point for the caravan routes which pass to MongoHa, Russia, Chinese 
Turkestan and regions farther west. 

The Chinese name of the town is Chang Chia-k'ou. The foreign name, 
Kalgan, according to Professor George B. Barbour, 1929, is probably a Russian 
corruption of the Mongol word khalga, meaning a barrier, or gate. The Great 
Wall blocks the entrance to the Ching Ho (Clearwater River) gorge, and 
behind it the town straggles along between the hills and the river, finally 
spreading out on both sides of the stream. The Great Wall, which was started 
in 214 B. C. by Shih Huang-ti, provides a formidable defense. Of Kalgan's 
history Barbour remarks: 

"The importance of the locality in Chinese history is clear from the long 
tale of military' engagements that took place from the earliest times in attempts 
to secure control of the district. It is recorded that in 128 B. C. General Wei 
Ch'ing marched up the Kalgan Pass to subjugate the city of Liang. In one of 
Genghis Khan's campaigns (121 2-1 213 A. D.) possession of Hsuan-hua, the 
walled city twenty miles to the southeast, was hotly contested and a bloody 
battle was fought at Huai-lai. . . . 

"The town itself is of relatively recent date. According to the Great 
Encyclopaedia of the Emperor K'ang Hsi, no city wall existed until the fourth 
year of the reign of the Ming Emperor, Hsuan Te (1429 A. D.). As it stands 
to-day — twenty-five feet high, twenty feet broad and four li in circumference — 
it dates from 1573, a fact known from the stamp impressed in bricks used to 
build it. 

"The other walled towns in the area date from the same dynasty. Wan 
Ch'uan, the next most important as guarding the. western pass, was built 
in 1392."' 

There are two passes from Kalgan to Mongolia, namely: Hanopa and Wan 
Ch'uan. The oldest, Hanopa, is so rough that when I took cars over it in 
19 1 8-1 9 19 they had to be hauled up by mules. In 1920 enough work had been 
done on the Wan Ch'uan Pass so that the road was passable for motors under 
their own power. Passable, but that is about aU, for it was such a mass of 
rocks and ruts that only the strongest car could endvue the pvmishment. 
Later, a heavy tax was imposed upon aU motors, and soldiers did enough 
road work to make a fairly decent highway as long as carts were prevented 
from using the automobile track. In the fortimes of war, Kalgan has changed 
hands with almost kaleidoscopic rapidity during the last five years, and the 
condition of the road up the Pass varies with each new general in control. 

' Barbour, George B. 1929. "The Geology of the Kalgan Area," Mem. Ceol. Survey oj China, Series A, 
No. 6, p. 3. 



We left Peking for Kalgan on April 17, 1922. I had been provided by the 
Chinese Ministry of Foreign Affairs with a formidable-looking dociiment, 
which was supposed to permit our cars and equipment to leave Kalgan exempt 
from duty and customs inspection. When I presented it to Chang Tso-lin's 
soldiers, who were stationed on the road to the Wan Ch'uan Pass, they laughed 
contemptuously and said: "This is from Peking. We don't recognize Peking." 

They insisted that a new permit be obtained from the local military com- 
mander. Although it could have been issued in ten minutes, it required three 
days of constant telephoning and visiting the yamen before it was forth- 
coming. In the meantime, I sent all oiu- heavy baggage by cart to the village 
of Miao-T'an, on the plateau, thirty-five miles from Kalgan. The Pass was 
reported to be in very bad condition, and it was imperative that otir cars 
shotdd go up as light as possible. 

Since the Mongols do virtually no farming or manufacturing, their sim- 
ple needs, such as saddles, cloth, tea, tobacco, boots, tents, etc., are supplied 
by the Chinese. In Kalgan or Kweihwating, farther to the west, one can pur- 
chase all the necessities of Mongol life. Caravans take these commodities into 
Mongolia and barter for hides, fiu-s, wool, sheep and ponies. From these two 
frontier cities, trade flows to and from central Asia as it has done for centviries 
and will continue to do until the desert is crossed by raUroads. 



We left the Anderson, Meyer and Company's compound in Kalgan in five 
cars at six o'clock on the morning of April 21 , 1922. Before we were out of the 
city gates, we were joined by two other motors. One was driven by Mr. Charles 
L. Coltman, who was en route to Urga on business. In the other were Mrs. 
Granger, Mrs. Shackelford, and Mrs. Black, who were going to the summit 
of the Wan Ch'uan Pass to see us safely on our way. I was taking Mrs. 
Andrews as far as Urga in order that she might get some Paget color photos 
of the brilliant Mongol costumes for the Expedition. Dr. Davidson Black, 
of the Peking Union Medical College, who had joined the Expedition tem- 
porarily in order to obtain data for his anthropological studies, was to retvim 
from Urga with Mrs. Andrews. 

The seven cars, finally under way, made a very imposing spectacle as they 
woimd up the long river-bed leading to the plateau. As far as the base of the 
Pass the country is deeply blanketed with the Pleistocene loess. Centuries 
of travel in the same cart-ruts have worn the roads into canyons which often 
lie many feet below the surface. The loess hills teem with human life. Whole 
villages are half dug, half built into the hillsides but are well-nigh invisible, 
for every wall and roof is of the same brown earth. These cave-dweUings are 
delightfully cool in summer and easily warmed diu"ing the winter; their 
greatest disadvantage is the lack of light. 

Passing the splendid old city walls of Wan Ch'uan, we reached the foot 
of the Pass, an altitude of three thousand and seventy-five feet, seven miles 
from Kalgan. From this point the ascent to the plateau becomes very abrupt, 
in many places rising three hundred feet to the mile. The highest point, five 
thousand and seven feet, is reached just where the road breaks through the 
Great Wall which stretches its length like an ancient gray serpent along the 
edge of the plateau. In eleven miles the trail rises nineteen hundred and twenty- 
five feet. 

From the Great Wall there is a superb view. Looking backward to the 
south, mile after mile of rolling hills stretch away to where the far horizon 




meets the blue haze of the Shansi Mountains. The waves in this vast land- 
scape have been cut and slashed by the agents of erosion: wind, frost and rain. 
The hills lie in a chaotic mass with gaping wounds and giillies, painted in rain- 
bow colors, crossing and cutting one another at fantastic angles as far as the 
eye can see: a stupendous relief map of desolate land. 

Forward to the north lies Mongolia, a seemingly limitless stretch of 
undulating plain swelling to roimded hills between shallow, cultivated valleys. 
One feels that this is the roof of the world and that beyond the next hillock 
one can look over the eaves to the rest of the earth spread out below him. One 
breathes the clean dry air of Mongolia: air so invigorating that it spvirs on the 
modem explorer even as it must have done his primitive ancestors. 


The slope to the margin of the plateau, up which we had come, is typical 
"badland" coimtry, a beautiful example of wind and water erosion. Profes- 
sors Berkey and Morris could hardly be restrained from plunging at once into 
the geological problem which it presented, but we covild give them time for 
only a cursory examination. The region was supposed to be Upper Jxirassic 
and barren of fossils, but the geologists were convinced, on structural grounds, 
of its post -Jurassic age. After ovir rettim from Mongolia they gave it fiirther 

Cambro- Lower Lower Terliar_y Pliocene & 

Ordovician Jurassic Cretaceous Lava Pieislocene 

Figure 3. — Generalized geologic section from Tientsin to the Wan Ch'uan Hsien Pass. From "Geology 
of Mongolia." 

The geologists have this to say about the entrance to Mongolia: "The 
edge of the plateau reached by the Wan Ch'uan Hsien Pass, twenty miles 
above Kalgan, is a true continental divide. From that point toward the inte- 
rior the average elevation gradually decreases, and one enters the great basin 
occupied by the Desert of Gobi. While the elevation at the divide is more 
than five thousand feet above the sea, at the center of the basin, three hundred 
miles away, it is less than three thousand feet. From the top of the Pass the 
streams turn inland, and for nearly six himdred miles the drainage is towards 
the interior."^ 

' Berkey, Charles P., and Morris, Frederick K. 
Central Asia, Amer. Mus. Nat. Hist., II, p. 49. 

1927. "Geology of Mongolia," Natural History of 


\ ioiVIDE 




|kiang pacific divide t 1. 

iO (00 


300 400 

900 600 MILLS 





^ profile: B 

Figure 4. — Two cross profiles of the Gobi Basin. From "Geology of Mongolia." Profile A crosses the 
whole of the eastern Gobi and shows a broad, shallow downwarp between the Arctic and Pacific Divides. 
Profile B crosses only the northwestern extension of the Gobi lying north of the Altais and shows much stronger 
warping and block-faulting than does the eastern section. 

Altogether there are nearly five thousand feet of exposed sediments. The 
lower three thousand feet consist of coarse pebbles, chiefly of trachyte and 
granite porph>Ties; the upper two thousand feet consist of finer sands and 
clays which contain fragmentary dinosaur bones. The geologists designated 
the deposit as the Wan Ch'uan formation. They determined with little doubt 
that the strata are early Cretaceous. Later, Professor George B. Barbour, of 
Yenching L^niversity, Peking, spent considerable time investigating the Kalgan 
area and has published the residts of his studies in the Mevioirs of the Geological 
Survey of China, Series A, No. 6, 1929. 


The Great Wall of China was built to keep the Mongols out and, by the 
same token, it should have kept the Chinese in, but rich ground for agrictdture 
has lured the Chinese farther and farther into the grasslands of Inner Mongolia. 
Thus, when our cars roared through the gateway of the Wall, we entered a 
farmland region dotted with the brown walls of Chinese mud villages where 
blue-clad peasants were already at work in the fields. 

We had had rough going in the Pass — ruts, rocks and steep climbs — but 
the mud beyond the Wall was our greatest diflficiilty. Subsequently, we 
learned that mud is the one thing we need fear for motor transport. When 
the terrain is dry and the wheels can grip, almost any obstacle can be sur- 
mounted. Even sand does not worry us tmduly, but wheels that slip and spin 
on mud are useless. The fifteen miles of road between the top of the Pass and 
the village of Miao T'an where ox\x baggage was awaiting us, were very bad. 
As a matter of fact they are almost always bad. The ground is low, and water 
drains off slowlj^ so that mud holes may remain for days after even a light rain. 

We reached Miao T'an at three o'clock in the afternoon and by foiu- were 
ready to leave. Bayard Colgate, Chief of Motor Transport, and I were horri- 
fied at the loads which were piled upon the cars. They consisted largely of 
gasoline, photographic equipment and automobile tires which had arrived too 
late to go with the camels and coiild not be left behind. Fortunately, the road 
was smooth and hard — better than I have ever seen it since — and we thought 


we might be able to get beyond the area of Chinese ctiltivation before dark and 
camp in the grasslands. 

Brigands are a very real menace in the Chinese region where the caravan 
trails converge toward Kalgan and it is highly dangerous to spend the night 
near one of the villages. In later years we learned to arrive at Miao T'an late 
in the afternoon and sleep there, so that we might have the whole day in which 
to travel the brigand-infested area. The experience which we had the first 
night, while uncomfortable, was valuable. It had begun to drizzle and there 
were still twenty-five miles to go. We were driving through the inky blackness 
of a rainy night when suddenly the cars ran into soft groimd and became so 
hopelessly mired that we could only camp where we were. We worked for 
many hours before they were extricated, but it taught us the lesson that we 
should never travel at night unless it was absolutely unavoidable. 


For one hundred and ten miles north of Kalgan, in 1922, there were fields 
owned and cultivated by the Chinese. The Chinese Governor of the district 
opens a new region for settlement every two or three years, and as soon as the 
land is sold to colonists more is taken. The average advance is about twelve 
miles each year. The Mongols are being pushed northward closer to the edge 
of the desert, for they do no farming and depend for their living upon pastoral 
pitrsuits. Millet, oats, wheat, barley and potatoes are the chief products of the 
Chinese farmers, and the yield is excellent since the land is very fertile. 

The members of the Expedition were disappointed that we saw few Mon- 
gols even though we were in Mongolia. As a matter of fact, until the area of 
Chinese cultivation is passed Mongols are virtually non-existent, for they have 
been driven out by the steady advance of Chinese agrictdtiire. Ten miles 
beyond the Tabool Hills the village of Chap Ser was formerly almost entirely 
Mongol, but to-day it has become a collection of Chinese mud huts. After 
1922, when motor cars had become an established means of transportation 
between Kalgan and Urga, Chap Ser was used as a gasoline station and was 
the first night's stopping place for commercial cars on the northward trip. 


AU through the ctoltivated region we saw great flocks of waterfowl mov- 
ing northward on their spring migration. Since ornithology was not a part of 
the Expedition's work, game birds were of most interest to us for they formed a 
welcome addition to our food supply. I wiU mention a few of the most abun- 
dant. In the fields the bean geese, Anser fabalis fabalis, were feeding by thou- 
sands on the young shoots of winter wheat. They were rather difficult to 
approach if one were on foot, but when in a cart or walking behind a pony one 


could get within easy shooting range. Although this is the most common 
goose in north China, strangely enough I have seen only one in the north 
Mongolian grasslands. There the bar-headed goose, Anser indicus, the swan 
goose, Cygnopsis cygnoides, and the graylag, Anser anser, breed in small num- 
bers, but the bean goose appears not to nest there. 

In the spring, the grasslands of Inner Mongolia are dotted with ponds, 
many of which disappear later in the svunmer. These form a convenient halt- 
ing place for ducks of several species, some of which remain to breed. The 
mallard, Anas platyrhynchos, and the teal, Anas crecca crecca, outnumber all 
the rest, but I have shot the falcated teal, Anas falcata, pintail. Anas acuta 
acuta, sho-veler, Spatula clypeata, red-head, Nyroca ferina ferina, and the tufted 
bluebtll, Nyroca fidigula, both in the spring and in the autumn, when we have 
been going to, or returning from, the desert. 

The ruddy sheldrake, Casarca ferruginea, is omnipresent even in the Gobi, 
and I shall tell of its peculiar habits in a futiire chapter. The beautiful burrow 
sheldrake, Tadorna tadorna, we found on almost all the ponds or lakes from the 
Chinese frontier north to Urga, but when breeding it does not stray far from 
water, as does the ruddy sheldrake. 

There were many species of shore birds on the grasslands, the most abim- 
dant being the greater sand plover, Charadrius leschenauUii, little ringed plover, 
Charadrius dubius curonicus, golden plover, Charadrius dominiciis fulvus, lap- 
wing, Vanellus vanellus, and gray-headed lapwing, Microsarcops cinereus. 
The two last-named species breed in Mongolia, and the former is one of the 
most common birds wherever there is a bit of marshland. Its melancholy cry 
never fails to bring to my mind a picture of the windswept reaches of the 
northern grasslands, where I have often watched it try to entice me away 
from the vicinity of its nest by simulating an injured wing. The ciirlew, 
Numenius arquata, and the avocet, Recurvirostra avosetta, are fairly common 
near ponds, and we could usually pick up a few snipe, Capella gallinago 
raddei, in wet ground. 

I must not neglect to mention the beautiful demoiselle crane, A nthropoides 
virgo, which is one of the most abimdant and beautiftd of the large Mongolian 
birds. Great numbers were circling in ascending spirals or feeding in the wheat 
fields beside the road as we passed. These cranes breed in northern and cen- 
tral Mongolia and they are so unafraid that a man can almost touch them, 
but when in flocks they are not easily approached. 


While we were in the area of cultivation it was difficult to do geological 
work; as the fields became less frequent and finally ended, rock outcrops 
appeared more often. A gently rolling plain continued until we reached the 


1 th( center of the piclure. At the Tight the Great Wall of China defendine the Past climbs o 


Tabool Hills, eighty miles from Kalgan; it varied in elevation little more than 
one hundred feet. The geologists Berkey and Morris, 1924, foimd the struc- 
ture to consist of very ancient granite and metamorphic rocks, which later 
they realized were a part of "The Great Mongolian Bathylith." 

The geologists had a most difficvilt and strenuous task to interpret the 
structiire correctly and still keep in touch with the rapidly moving main 
party. I think that none of us realized just what we were asking them to do. 
They had been plunged into the midst of extremely complicated problems in 
an unknown coimtry. A correct interpretation at the beginning and a continu- 
ous unbroken record of structiiral changes were of the utmost importance to 
their futtire work. They had to solve these problems, at least partially, and 
preserve the integrity of their record "on the wing" as it were. They could 
remain for only a few moments at any one spot unless it was of the utmost 
importance, since it was unwise for them to lag too far behind the motor fleet. 
The type of men they were is evidenced by the fact that they did not complain 
but plunged into their well-nigh impossible task with the greatest enthusiasm 
and energy. 

It was necessary to make literally thousands of observations, inspections, 
judgments and trail studies as rapidly as it was possible to work. To record 
these observations they kept a route map on which were sketched the course 
of the traverse and the bordering topography. From aneroid barometer read- 
ings a running profile was constructed, mile by mile. On this the geology was 
sketched in cross sections, representing the imdergrotmd structure, the succes- 
sion of formations and the interpretation of their relations one to another. 
When the Expedition was moving rapidly, such an undertaking was extremely 
trying and exhaustive work, yet sometimes they did more than one hundred 
miles of such cross section studies in a single day. (See Figures 5 and 6.) 


Even though it was evident that in certain areas thin deposits of sediments 
overlay the hard rock floor, there were no promising exposures and we did 
not attempt palseontological investigations. Moreover, in 1919-1920, Dr. J. G. 
Andersson had done some prospecting in this same region while spending the 
summers at the Swedish Mission of Hallong Osso, only a few miles from Tabool. 
He had discovered several smaU fossil deposits, the largest of which was in an 
area called Ertemte. While digging a well, the Mongols had come upon a 
sandy layer at a depth of three meters, containing a fossil fauna consisting 
largely of rodents. Most interesting among the specimens was the molar of a 
large beaver-like rodent. At several other near-by localities Doctor Andersson 
found smaU deposits, either in well-diggings or in the banks along dry stream- 


Doctor Andersson has remarked, "Because of the almost complete lack of 
natural sections in the gently undulating Mongolian steppe it is very difficult to 
determine the relation and age even between localities situated close together."^ 

Nevertheless, he provisionally identified two Pleistocene deposits con- 
taining Elcphas and Rhinoceros, and Pliocene and late Miocene horizons in 
several localities. Among the fossils at Ertemte, which he assigned to the 
Pliocene or late Miocene, were Castor and other rodents, Cervavus and Stru- 
thioUthus. An earlier horizon, also Pliocene or late Miocene, was characterized 
by the presence of "Aceratherium, Hipparion and Artiodactyla, the whole 
closely resembling the Hipparion fauna of China proper."^ 

Speaking of Ertemte, Doctor Andersson remarks: "When in 19 19 I 
thought that I had identified a beaver molar in this deposit, an identification 
which has since been fully confirmed, I felt I was facing a palaeo-climatic 
problem of singular interest. 

"At present the Hallong Osso region is a typical steppe, where trees are 
exceedingly scarce, occurring only in small groups in some sheltered rock 
ravines. Otherwise the whole area is an open rolling grassland. There are 
no permanent watercotorses, and the lakes are more or less saline and entirely 
dry a large part of the year with a thin crust of salt, which is replaced by water 
only after occasional rains. 

"The finding of fossil beaver, a type indicating forests and water, cer- 
tainly invites continued research and thoughtful consideration. . . . 

"If the Ertemte sand was deposited in an ancient lake, it means, as 
clearly shown on the map, that the climate of that period was immensely differ- 
ent from that of the present time, for this region is now characterized by 
treeless vegetation and small residual salt lakes. "^ 


Although there are several roads to Urga, the one which we were following, 
and which the cars all use, is bordered by the telegraph line and is commonly 
referred to as the "Telegraph Road." It is generally in fair condition except 
when there has been an imusual amoimt of cart traffic, which leaves deep ruts. 
Because of the war between the Chinese and Mongols in 1921, no carts had 
since used the road, and when we had passed Chap Ser it was like a boulevard. 
It wound among rounded hills tinged with the first green of early grass, and 
our cars, in spite of their heavy loads, roared on without an accident. 

Ovir second camp was made at five o'clock in the afternoon, in a beautif\il 
amphitheater where the hills rolled away in gentle yellow-green waves from the 

' Andersson, J. G. 1923. "Essays on the Cenozoic of Northern China," Mem. Geol. Survey of China, 
Series A, No. 3, p. 46. 
>0p. «V., p. 47. 
» Op. cit., pp. 49-52. 


granite rocks behind the basin. The early camp gave us time to find necessary 
items of equipment and to eliminate some of the mud of the previous night. 
The tents went up like magic, and in half an hoior a tiny city appeared in the 
grassy valley. 


Within a week the Expedition had been so organized that making and 
breaking camp were short operations. When I had designated the spot for the 
mess tent, which was always the center of the camp, the foreign staff arranged 
their tents on one side, and the natives on the other. Every man knew exactly 
what he had to do, and it was seldom longer than twenty minutes from the 
time the last car came into camp until all the tents were up and the cooks had a 
fire started. Then each man set about the tasks pertaining to his particular 
function on the Expedition. The motor experts refilled all the tanks with gaso- 
line and oil and made a thorough inspection of every car, tightening bolts and 
nuts, patching tires and putting all in readiness for the next day's run. A 
record of the mileage, of gasoline and oil consumption, of leakage, and of the 
terrain was kept by the Chief of Motor Transport. 

The geologists set to work at once transcribing their notes of the day's 
run into more permanent form. Often this carried them far into the night, 
and it was seldom that the light in their tent was extinguished before twelve 
o'clock. The photographer reloaded his motion-picttire magazines and 
recorded the subjects photographed during the day. If we were near a prom- 
ising-looking exposure, the palaeontologist made a rapid search for fossils. 

I usually despatched the three Chinese taxidermists in different directions 
to set traps for small mammals, and often Granger and I went with them. 
Even though we might catch only ten or fifteen specimens, it gave us a con- 
tinuous faunal line over the region which we were traversing, and often served 
to delineate the limits of distribution of a species or genus. 

The evening ho\irs about the mess table were those that I remember most 
vividly. Each man had a different viewpoint on the country over which we 
traveled, and to have such a fvmd of expert knowledge in so many branches of 
science upon which to draw was like reading a condensed volume of the "Book 
of Knowledge." Thus, within a very few days after leaving Kalgan, we had 
developed an organization of camp life in which the Expedition operated like 
a well-oiled machine. Moreover, a superabundance of interesting work, good 
food and congenial personalities contributed to make us a "happy family." 


After leaving our second camp we ran for thirty miles over pleasant roll- 
ing grasslands, where I had promised my companions a glimpse of antelope. 


They were mildly skeptical about my tales of the sixty-miles-an-hour speed of 
gazelles and I prayed for some which would give an exhibition of really high- 
class running. We did find a herd, but the country was so rough that we could 
not push the animals sufficiently to make them really run. However, it was 
sufficient to turn the doubters into my firm supporters. 

I killed a young buck, and my companions examined the animal with the 
greatest interest, for it was their first sight of a Mongolian antelope. The 
species was the goitered gazelle, Procapra gutturosa, which is confined to the 
grasslands and seldom ranges into the arid parts of the Gobi. The larger size, 
short tail, heavier body, shorter legs and smaller horns distinguish this animal 
at once from the desert gazelle, Gazella stibgiitturosa hiUieriana. These are 
the only two species of gazelle in eastern and central Mongolia. 

The goitered gazelle is so named because of a great enlargement of the 
larynx, which produces a prominent swelling on the throat. WhaX purpose 
this extraordinary character serves I am at a loss to know. Certainly it is not 
to give them an exceptional "voice," for, when wounded, I have heard them 
make only a deep-toned roar. During the winter the Mongolian antelope 
grows a coat of very long, soft hair which is light brown-gray in color, strongly 
tinged with rufous on the head and face. Its summer pelage is beautiful orange- 
fawn. The winter coat is not fully shed until the end of May, and the skin of 
the buck I killed in April was useless as a museum specimen. 

Ten years ago gazelles were abundant at Tabool and farther south to 
within fifty miles of Kalgan. Like the Mongols, they have been driven from 
their former range by the steady advance of the Chinese farmer; they may 
usually be found, however, within five or ten miles of the cultivated area. 
Both species of gazelles were of prime importance to our Expedition because 
we depended upon them for fresh meat. Fortunately their flesh is so delicious 
that we never tired of it, even though hardly a meal passed that antelope meat 
in some form was not our principal dish. I shall reserve a discussion of their 
habits for future chapters. 

Except for the antelopes, the only visible wild animals in these southern 
grasslands were the spermophiles, Citellus dauricus mongoliciis, and occa- 
tionally a kangaroo rat, AUactaga mongoUca annulata. Wolves and foxes are 
natives of this region but they are not abundant ; the little field vole, Microtiis 
warringtoni, has been caught near Tabool. 

The spermophiles, or groimd squirrels, were abimdant and I shot eighteen 
or twenty from the car. This species has a wide range in north China and 
Inner Mongolia, and was originally described b}^ Milne-Edwards in 1867 from 
specimens taken near Peking. It is confined to the grasslands, and as we 
approached the edge of the desert near P'ang Kiang it disappeared. 

The beautiful kangaroo rat is chiefly nocturnal, but we sometimes saw it 



Professors Berkey and Morris carried their geological sections in this way throughout the entire route 

of the Expedition. 

B. AX L\rLDIll()\ CAR i;\ i>li- i- II ULTIES, I928. 
Often an apparently hard crust would break, letting the car sink into soft, sticky mud. 




^ - 

< i 
-= i 

< w 

O " 

2 « 


< a 


running beside the road in the late afternoon. By means of its enormous hind 
legs it is able to jump six or eight feet and can easily distance a man. The 
M'ongol name for all jerboas is Allactaga, which has been adopted as the desig- 
nation of the genus. The particular species, Allactaga mojigolica annulata, 
which we saw near Tabool in the grasslands also ranges southward into north- 
eastern China. In the Central Gobi it gives place to the somewhat lighter- 
colored species, Allactaga mongolica. Ancient Mongolian legends describe 
the allactaga as the horse of the hero Tarabagan, which was transformed into 
this leaping rodent. The hero himself was changed by his enraged god into a 
marmot. Thus the Mongol name for the marmot is "tarabagan." Evidences 
of plague have been found in the Allactaga, and natives rather avoid the 



As we approached P'ang Kiang, the first telegraph station on the road to 
Urga, the country began to change very gradually. The grass became thinner 
and coarser, and small clumps of camel sage appeared. 

The geological structiure changed as weU. We had come out upon what 
ovir geologists have named the "Gobi erosion plane," a surface of extraordinary 
smoothness developed upon the relatively soft basin sediments. These low 
erosion plains, so characteristic of Mongolia, are the resvilt of the infinitely 
slow, but irresistible, forces of weathering which have worn down the moim- 
tains and carried off the waste, leaving a new plain carved from a former 
mountainous region. The Gobi erosion plane is interrupted by many vmdrained 
hollows which range in size from two himdred yards to tens of miles in length, 
and from twenty feet to four hundred feet in depth. The larger hoUows have 
relatively flat floors, though they are never so perfectly level as the Gobi 
upland, and in almost all cases contain several shallow playas, or wet-weather 


Fifteen or twenty miles before we reached P'ang Kiang the plain began 
to be dissected by ravines and gullies, and red hills and buttes showed promi- 
nently against the skyline. It was ideal countrj^ in which to search for fossils 
and, in order to give the palaeontologist an opportvmity to inspect the expo- 
sures, we camped on the plain not far from the telegraph station. 

Berkey, Morris and Granger spent the afternoon visiting the most prom- 
ising localities, but discovered only one fossil, a fragment of a rodent jaw. 
Doctor Matthew identified it as an ochotonid, which is not sufficient to serve 
as an index of the exact age of the formations. In 1928 the deposit was iden- 
tified by Doctor Spock^ as Pliocene. These P'ang Kiang beds, where exposed 
sixty miles farther south, are about five hundred feet thick and, in some places 

'Spock, L. Erskine. 1929. "Pliocene Beds of the Iren Gobi," Amer. Mas. Novitates, No. 394; Pre- 
liminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., II, No. 96, pp. 1-8, 1930. 










- w 



at least, rest directly upon the old crystalline rocks. It is at P'ang Kiang that 
we always look for "Gobi stones," those polished nodules of chert, agate, 
chalcedony and jasper which are beautiful in color and shape. 

P'ang Kiang boasts of only four mud huts, most important of which is 
the telegraph station. An inn for the convenience of motor-passengers was con- 
structed in 1923 and for two years did a flourishing business with Chinese 
merchants traveling between Kalgan and Urga. A more desolate spot hardly 
can be conceived. The buildings are situated in the lowest part of the basin, 
their original location being dictated by the presence of a well containing some 
of the worst water in Mongolia. After enduring this for several years, the 
Chinese dug two other wells which produce fairly good water, only moderately 
impregnated with alkali. The Chinese telegraph operator, who had been at 
this station for seven vears, managed to keep quite contented with the assist- 
ance of opium. 

The "City of P'ang Kiang," as it was often referred to in Chinese papers, 
had been the scene of important events since I last visited it in 191 9. After 
the Mongols and Russians drove the Chinese out of Urga, they carried the war 
into Inner Mongolia and for several months P'ang Kiang was the first line of 
Chinese defense. The long hill slope opposite the telegraph station was pitted 
with large, horseshoe-shaped depressions reinforced with concrete and arranged 
in regular lines. These were the "basements" of the quarters in which the 
Chinese soldiers lived during the long winter of 1921. 

P'ang Kiang marks the southern edge of the Gobi Desert, on the Kalgan- 
Urga trail, if the Gobi can properly be said to have an "edge." The grasslands 
merge so gradually into the arid regions of the Gobi that it is difficult to say 
just where the real desert begins. The thinner and coarser grass, the presence 
of sage and the terrain of fine gravel indicate the approach to an arid region. 


At P'ang Kiang one leaves the Southern Mongolian Life Zone character- 
ized by the fauna and flora of the grasslands and enters what I have called the 
Gobi Life Zone, which is that of a true central Asian desert. There is, of course, 
a certain overlapping of faimas, but the change is fairly abrupt. 

The short-tailed goitered antelope, Procapra guttiirosa, begins to give place 
to the long-taUed desert species, Gazella siibgutturosa hillieriana, which does not 
range south of P'ang Kiang. Not far from the telegraph station I have seen 
them both together in the same herd, but that is very tinusual. 

The grassland spermophile, Citellus dauricus mongoliciis, does not reach 
P'ang Kiang and apparently no other species of the genus takes its place. 
This leaves a break of two himdred miles in the north-south range of the 
genus, where I have seen no spermophiles during many trips across the Gobi, 


A pale-colored jerboa, Allactaga mongolica, replaces the darker Allactaga 
mongolica annidata just beyond P'ang Kiang, and a desert hamster which Dr. 
G. M. Allen, 1925, has described as a new race, Cricetulus migratorius, appears 
in abundance. 

The lizards, Phrynocephalus cf. versicolor and Eremias przewalskii, typical 
desert species, scurry about the gravel plain, and the brown pit-viper, Agkis- 
trodon halys intermedins, which is the only poisonous snake that we have 
encountered in the Gobi, is fairly abundant. 

The Gobi Life Zone is, of course, less well-defined by the avifauna, but one 
typical species, the sand-grouse, Syrrhaptes paradoxus, seldom appears south of 
P'ang Kiang except during the winter when the extreme cold drives it into 
northern China. 

The chough, Pyrrhocorax pyrrhocorax, and the red-leg partridge, or 
chukar, Alectoris grceca pubescens, are both inhabitants of the rocky hills of 
Inner Mongolia but stop short of the desert at P'ang Kiang. The great bus- 
tard, Otis tarda dybowskii, appears to prefer the grassy plains rather than the 
more arid regions, but we found it pretty generally distributed all over Mon- 


After leaving Chap Ser on the way to P'ang Kiang, we had passed through 
the districts of the Chahar and Sionit tribes of the true southern Mongols. 
Although the natives nearest the frontier of China have adopted Chinese cus- 
toms to some extent, still I believe that there is very little intermarriage 
between the two races and that most of the southern Mongols are of pure blood. 

They live in the dome-shaped felt yurts which I have described in Chapter 
I (p. 16) and, so far as I have observed, have much the same habits as the 
more northern natives, except that they are by no means as nomadic. It is 
not necessary for them to move so frequently because there is usually sufficient 
grass in a single locality to last their flocks through the stimmer. As a resvilt, 
whole villages will remain in the same place year after year. This is seldom 
true of the desert Mongols, who have a regular cycle of migrations. The 
southern natives, being near an agricultural region, eat much more flour and 
more vegetables than do the desert people; the latter live almost entirely 
upon animal products. 

The most distingmshing feattire of the Chahar and Sunit Mongols is the 
headdress of the women. This is a network of red coral and silver, with long 
pendants of coral, silver and turquoise which reach below the shoulders. It is 
strikingly trnlike the Khalkas and other northern tribes where the women dress 
their hair over a great framework like the horns of a mountain sheep. Although 
all of the Expedition Mongols were from the Chahar region I have not lived 



with the southern tribesmen as I have done with the northerners, and therefore 
do not feel qualified to compare the details of their home life. 


The Expedition camped only one night at P'ang Kiang. On the fourth 
morning we had an early start for Iren Dabasu (Erhlien of the Chinese), the 
second telegraph station on the Urga road, where I had instructed otu* caravan 
to leave us a supply of gasoline. Twenty miles from P'ang Kiang we passed a 
small temple which stands isolated a short distance from a well of excellent 
water. The place was deserted, and scattered about the plain were dozens of 
soldiers' imiforms and lama robes, some of them containing weathered human 
bones. Pariah dogs — grim evidence of the fate of the unfortunate dead — 
slunk in and out of the gaping walls. It was eloquent testimony of the war 
which had ravaged this once peaceful land. 

As we proceeded northwestward toward Iren Dabasu and penetrated 
deeper into the desert, the country became more and more arid. There was 
little vegetation except small tufts of hard stiff grass, camel sage and a low 
thorny bush about eighteen inches high. We were traveling over a vast plain 
of Tertiary sediments which, being undissected, offered no prospects to the fossil 
hunters. Twenty-three miles before we reached Iren Dabasu the road dropped 
abruptly off the gravel plain, which is almost as level as a table, into a vast 
basin, in the center of which lies the telegraph station, near a partially dry 
salt marsh. 

The descent to the basin floor was down a steep bluff, the face of which 
had been deeply eroded by wind and water. It was obviously a place which 
must be investigated by Berkey, Morris and Granger, so I told them that I 

Figure 7. — Tlie scarp at Irdin Manha. From "Geology of Mongolia." 

would run on to make camp at Iren Dabasu and leave them to follow more 
slowly. Five miles from the telegraph station we descended another, smaller 
bluff, into the bottom of the depression which marks the lowest spot in the 
great Gobi Basin between Kalgan and Urga. The altitude is only twenty-nine 
hundred and thirty feet. 


We ran over to the telegraph station, which consisted of two Chinese mud 
houses, found our cache of gasoHne and learned that our caravan had passed 
that way two weeks earlier. Then we drove to some gray-white ridges, half a 
mile to the west, and pitched the tents. 


We were hardly settled before the last two cars swung aroimd a brown 
earth bank and roared into camp. The men were obviously excited when I 
went out to meet them. I knew that something unusual had happened because 
no one said a word. Granger's eyes were shining and he was puffing violently 
at his pipe. Silently he dug into his pockets and produced a handful of bone 
fragments; out of his shirt came a rhinoceros tooth, and the various folds of 
his upper garments yielded other fossils. Berkey and Morris were loaded in 
a similar manner. Granger held out his hand: "Well, Roy, we've done it. 
The stuff is here," he said. We were very happy. It was the first definite 
evidence that our exploration in Mongolia, so far as fossils were concerned, 
would not be the fruitless hunt predicted by our friends in Peking. 

The specimens had been f oimd in the second blufE only five miles from camp 
and hardly twenty feet from the road. The horizon was certainly Middle 
Tertiary, but the bones were too fragmentary to permit of a positive identifi- 
cation except in the cases of an aquatic rhinoceros, Cadurcotherium, and of a 
large entelodont. While dinner was being prepared. Granger wandered off 
along the gray-white ridge that lay like a recumbent reptile west of camp. 
Even in the failing light he found a few fossil bits and we realized that we had 
another deposit at our very door. 

The following morning, before breakfast. Doctor Berkey discovered a 
section of the femur of what appeared to be a dinosatir, on the ridge just above 
camp ; a few minutes later Dr. Davidson Black found part of the missing seg- 
ment in front of his tent. When, after a short time. Doctor Berkey came 
into camp to ask me to accompany him to the summit of the gray ridge, my 
excitement had risen to high pitch. Granger was at work exposing the tibia 
of a large dinosaur; there cotild be no doubt of it this time! 

"This means," said Doctor Berkey, "that we are standing on Cretaceous 
strata — the first Cretaceous and the first dinosaur known in eastern Asia!" 

Thus, foiu- days after entering Mongolia and within two himdred and 
sixty-five miles of Kalgan, we had discovered Cretaceous strata as weU as 
fossil-bearing Tertiary deposits. All this in a region which we had been told 
would yield us nothing! Can one be surprised that we were jubilant? 

It was not only a personal triumph for Berkey, Granger and Morris, but 
also for American science, for other geologists had traversed the same forma- 
tions but had failed to determine the strata correctly. Obruchev, the Russian 


geologist, had found a tooth of "rhinoceros" (possibly a titanothere) at the 
escarpment twenty-three miles south of Iren Dabasu, and took for granted 
the Tertiary age of the rest of the sedimentary beds. 


Granger and the geologists remained for a week in 1922 at Iren Dabasu to 
make a reconnaissance of the deposits; in 1923 they gave it a more careful 
study. The Iren Dabasu basin, as it is to-day, is a most desolate region. 
Spotted by conical sandy movmds, sparsely covered with thorny bushes and 
camel sage, it is a burning desert imder the summer sun and an arctic waste in 
winter. In the lowest part lies a marsh which is sometimes covered by a thin 
sheet of muddy water, but usually gleams whitely with encrusted salt. Along 
the edges Chinese laborers, recruited in Kalgan, collect the salt and send it 
southward to the markets in great caravans of oxcarts. That is done when 
there is no war. At the time of our visit in 1922, the only human life centered 
about the mud huts of the telegraph station and the motor inn. Three wells 
in the basin, which furnish the only water available for miles, have long made 
it an established camp for caravans. A few miles to the eastward stands a 
lama temple, but, like the one near P'ang Kiang, it was deserted and partially 

Iren Dabasu, with all its vmattractive surface svirroundings, held for us a 
veritable wealth of scientific treasure. I cannot do better than quote what 
Doctors Berkey and Granger have said, since it gives the key to the forma- 
tion of the whole Gobi Desert region : 

"Central Mongolia is structurally a series of later sedimentary basins 
tmderlain by a floor of more ancient rocks. These basins of later sedimentary 
strata are separated one from another by stretches of open groimd where these 
same ancient floor-rocks form the surface. This floor has a very complicated 
structure, its members ranging in age from Archaean to Jtirassic time. In mid- 
Mesozoic time, after repeated earlier mountain folding and extensive igneous 
activity, the region was peneplaned, and it is this old peneplane svuface that 
has been warped and faulted to make the basins which hold the sediments of 
later age, now proven to be exceptionally good fossil groimd."^ 

The Iren Dabasu basin measures sixty-three miles across from northwest 
to southeast. We know its limits north and south, and that it extends west 
more than a hundred miles ; it covers an area of at least ten thousand square 
miles and is one of the largest sedimentary basins of Mongolia. On the north 
it is bounded by slate hills, on the west by complex old crystalline rocks and 

' Berkey, Charles P., and Granger, Walter. 1923. "Later Sediments of the Desert Basins of Central 
Mongolia," Amer. Mtis. Novilales, No. 77, pp. 1-3; Preliminary Reports, Central Asiatic Expeditions, Amer. 
Mus. Nat. Hist., I, No. 7, pp. 1-16, 1926. 









ON A camel's FOOT. 


on the south by hills of granite, schist and grajrwacke. It is floored with ancient 
slates and limestones. 


The horizontal Cretaceous beds lie immediately on the slate floor of the 
depression and have a total thickness of not more than one hundred and eighty 
feet. Even in the few days which were allowed for the first inspection, dino- 
saurs of the duck-billed, bipedal iguanodont type, small running dinosavirs 
such as the ostrich-like Ornithomimiis, crocodiles and turtles were identified. 

This, the first important fossil-bearing deposit which we had discovered 
in Mongolia, was formally designated by the geologists as the Iren Dabasu 
formation of Lower Cretaceous age. The geologists found that this stratum, 
of the Age of Reptiles, was not the only fossiliferous bed in the great depression. 
Five miles south of the salt marsh, the continuation of the second escarpment 
which we had descended proved to consist of rather coarse yellow gravel. 
This contained fragmentary bones and teeth of an interesting water-loving 
rhinoceros, Cadurcotherium , and of an enormous pig, Entelodon dims; the 
calcaneum of a huge beast which Granger at once suspected to represent 
the aberrant rhinoceros, Baluchitherium, first discovered in Baluchistan, India, 
191 1, by my friend, C. Forster Cooper of Trinity College, Cambridge; also the 
remains of a large tortoise, Testudo. 

The fauna indicates that the gravels are of Lower Oligocene age. The 
beds are from fifteen to fifty feet thick. They are evidently the result of 
stream action and the fossils, having been rolled about by swiftly flowing 
water, are usually in a fragmentary condition. The geologists named these 
beds the Houldjin formation, from the local Mongol term for the escarpment. 

The first escarpment, twenty-three miles to the south, still in the same 
structural basin, proved to be one of the most interesting and productive of all 
our fossil fields. The formation represents the LTpper Eocene and has been 
given the Mongol name of Irdin Manha, the "Valley of the Jewels," which 
contains many translucent quartz pebbles. In addition to several genera 
and species of camivora, the fauna includes lophiodonts — small hoofed mam- 
mals that were allied to the tapirs on the one hand and to the rhinoceroses on 
the other — in great abundance, and, most interesting of all, positively identi- 
fiable remains of titanotheres. This fulfilled a brilliant prediction of Professor 
Osbom that we would find titanotheres in Central Asia. 

Doctor Granger and the geologists had only a week in which to make their 
stirvey of the Iren Dabasu region. It was of such importance that we intended 
to give it further study, of course, but our first year's work was to be strictly 
a reconnaissance, to explore and appraise the coimtry with a view to future 
investigation. More intensive work was done in 1923. 


There were still diplomatic arrangements to be made in Urga, and we 
must connect with the caravan which I had directed to await us at Tuerin, a 
lama monaster>' two hvmdred and fifty miles away. After remaining one day 
at Iren Dabasu, we left two complete units with the geologists and, on Tues- 
day, April 26, started for Tuerin with the other three cars. 


About four miles from Iren Dabasu is the boundary between Inner and 
Outer Mongolia, marked where the road crosses it by a large obo, or conical 
pile of stones, and a sign in Chinese and Mongol. Sixty-eight miles beyond 
the frontier is Ude, the third telegraph station on the road. When I first went 
to MongoUa, Ude consisted only of the telegraph office and one or two other 
mud houses; in 1928 it boasted a motor inn, several Chinese shops and two 
yamens or official posts. Passport examination and customs inspection are con- 
ducted by the Buriat officials, according to a system inaugurated by the Soviet 
"advisers." In 1922, when the Soviet influence was only beginning to make 
itself felt, the inspection was nominal and quickly ended; now, to penetrate 
Outer Mongolia through the political barriers is exceedingly difficult and, once 
in, it is even more difficult to get out. 

The telegraph station at Ude is situated under the lee of a rough granite 
area, representing an exposed part of the Great Mongolian Bathylith. It is 
just in the center of the Gobi Desert, on the Kalgan-Urga traU, being almost 
equidistant from P'ang Kiang on the south and Tuerin on the north, where 
the desert begins to merge into the northern grasslands. The surroundings are 
even more desolate and arid than Iren Dabasu, but, unlike that region, hold no 
interest for fossil-hunters. 

Before the present Mongol Government maintained a frontier guard at 
Ude, the rough granite hiUs, to the north, through which the trail runs were a 
favorite place for bandits to await passing caravans. In 191 8 I had a lively 
fight at this same place. Fortimately we were able to handle our rifles more 
effectively than the attackers, who, nevertheless, shot much too close for 


Forty miles beyond Ude otir geologists discovered a great deposit of 
Jurassic strata, which is the youngest of the sedimentary groups forming the 
old floor of the Gobi region. It consists of a series of conglomerates and sand- 
stones of continental t>-pe, simply folded or sometimes block-faulted, and 
quite free from great metamorphism. Similar rocks were foimd at several 
widely separated places, and the entire series consists of stream deposits; 
nowhere is there e\'idence of marine conditions. The strata are of great 


thickness: in one place as much as twenty-five thousand feet. Much to our 
disappointment, the only fossils which the geologists were able to discover in 
this series were a few plant remains, chiefly stems which were poorly preserved. 


Our heavily loaded cars did not negotiate the road without some difficulty. 
In several dry stream-beds we encovintered sand which gave us trouble. How- 
ever, we never considered sand to be a very serious obstacle if it was not too 
extensive in area; short patches merely necessitated hard work in lightening 
the cars, digging a trail and pushing, by all concerned. We carried wide 
canvas strips which were laid down in the worst places, but they did not prove 
to be a great success. Later we discovered that canvas reinforced by trans- 
verse pieces of rope was excellent. 

After leaving the granite outcrops beyond Ude, the cars ran rapidly up 
and down a succession of low hills and descended into a vast basin. From the 
rim we appeared to be gazing across an ocean; not the smallest hiU or rise of 
ground broke the line where earth and sky met in a faint blue haze. Our cars 
seemed like tiny boats in a limitless brown sea. It is more than sixty miles 
across the depression, but there is a convenient halting place almost in the 
center at two wells named Sain Usu (Good Water). 


The presence of wells makes it possible to travel in Mongolia. The coun- 
try is criss-crossed by caravan trails, and along the most important routes 
wells have been dug every ten to forty miles. Some of them are hiondreds of 
years old, for these camel tracks across Mongolia are among the most ancient 
in the world. The Mongols themselves lead such a nomadic life that their 
wanderings have taken them into almost all parts of the desert ; if they wish to 
remain at a spot for some time they dig a well. The water-level is not far below 
the surface, as a rule. If a proper place is selected in the bottom of a drj.' 
stream-bed or where a dip in the strata directs drainage to a certain point, one 
seldom has to dig deeper than twelve or fifteen feet ; often not so much. The 
wells usually are walled with stone or sod, and those along the main caravan 
trails are kept in fair condition. The water is surprisingly good; only in a few 
cases was it so strongly impregnated with alkali as to be unpalatable. On the 
hottest day of summer the water temperature would seldom exceed fifty 
degrees Fahrenheit, and often it was much less; in a weU fifteen feet deep I 
foimd ice along the sides on August 15. We seldom boiled the water for 
drinlving unless the drainage was bad or a dead animal was too near for safety. 

Beside most of the weUs there is a wooden trough for convenience in water- 
ing stock ; in those which are in frequent use a bucket made of skin attached to 


the end of a pole is often to be found. On otir expeditions the water problem 
never worried us, for we were certain to find low ground or dry river-beds 
where we could dig a well for ovirselves if it became necessary. In many parts 
of the Gobi irrigation would be a simple matter. 

We were svuprised at the great number of ponds visible from the road 
after leaving Sain Usu. There had been an unusual amount of snow the pre- 
vious winter (1921-22) and this superabundance of water was the result. Most 
of the ponds had disappeared by the end of May. 


After leaving Sain Usu we saw a great many gazelles of both species, and 
three times they were mixed in a single herd. The grassland species covild 
easily be distinguished, by its larger size and lighter color, from the desert 
form; in the remains of their winter coats they looked almost white. I do not 
believe that they ever interbreed, for among hundreds of specimens which I 
have shot I never saw one which gave indication of mixed characters. Although 
the grassland species sometimes ranges into the less arid parts of the Gobi, 
the desert gazelle apparently never goes into the grasslands. 

Among the slate rocks on the north side of the Sain Usu basin we saw a 
good many gophers, Citellus paUidicanda. As I have remarked, between 
P'ang Kiang and Ude there appears to be an area in which spermophiles are 
absent. There are other apparent gaps in the range of the genus between Ude 
and Sain Usu, and between Sain Usu and Tuerin. Careful collecting might 
show that these are not actual gaps in the range, but only areas where, for 
some reason, the animals are very scarce. 


Tuerin, where our cavaran was to await us, stands in the center of a vast 
plain. A ragged mass of granite rocks which rises high above the surrotmding 
country is known as the "Tuerin Mountain." As a matter of fact, the "moun- 
tain" is an exposure of the Great Mongolian Bathylith and represents the base 
of an ancient movmtain range, the upper parts of which have been worn away 
by erosion. It continues some distance to the west in a lower ridge. Forty 
miles before reaching Tuerin we could see the granite summits standing sharp 
and clear against the skyline. The peaks themselves are not more than three 
hundred feet in height, but they rise from a rocky base some distance above 
the level of the plain. 

We came to the motmtain just before noon and saw a great caravan 
camped beside the road. As we drew near, I made out the American flag • 
flying from one of the loads and recognized our boxes. It was our own caravan. 
Merin, the leader, said that they had arrived only an hour before. They 


started from Kalgan March 21, 1922, and we met them April 28, 1922. I had 
told Merin five weeks before to be at Tuerin on, or before, this day. 

It was a great relief to reach the caravan at Tuerin so that we might 
transfer to it automobile tires and other heavy equipment not in daily use; 
also to get a new supply of food, for we had left almost all of our provisions 
with the geologists at Iren Dabasu. 

As a matter of fact, we maintained just such close coordination of the 
caravan and motor party throughout the entire season. The camels were in 
good condition at the start. We realized after a short experience that, allowing 
for the necessary time to feed, they could not average much better than ten 
miles a day when on a march of several himdred mUes; on short marches they 
could be pushed to twenty or twenty-five miles in a day. Using this standard 
we could usually figure pretty accurately how long it would take them to 
reach a certain point, except when the maps were in error, as frequently hap- 
pened. It became a point of honor, however, for the Mongols to bring the 
caravan to the rendezvous on the appointed day and in not a single instance 
were they late. We did not have such good success in subsequent years when 
we were using camels that were in poor condition after several seasons of hard 
work, and when political conditions had become more difficult. 

Instructing the caravan to remain where it was vmtil I had selected a 
camp site, we ran over to the telegraph station some distance away. There I 
found a letter from Mr. F. A. Larsen, addressed to "Roy Chapman Andrews, 
Anywhere in Mongolia." Larsen had been conducting negotiations with the 
Mongol Government on our behalf, and reported that conditions were favor- 
able for our work but that I must come to Urga before we started west. 
The letter had been brought to Tuerin by Mr. K. P. Albertson, who had been 
in Urga for some weeks negotiating for the reconstruction of the telegraph 
line. During the recent fighting the line had been destroyed beyond Iren 
Dabasu and there was no communication with Urga except by letters carried 
by private individuals. The Tuerin telegraph station is the fourth, and last, 
between Kalgan and Urga, and lies just outside the rocks near a well of excel- 
lent water. 

We pitched the tents in a beautiful amphitheater walled on three sides by 
serrated peaks; then I sent Colgate to bring the camels. Shackelford had all 
his arrangements completed for motion pictvires and it was certain that we 
wovild never find a more picturesque spot in which to photograph the caravan. 

Whenever the caravan arrived at a camp it gave me a thrill. That day I 
saw it in motion for the first time and it was a sight I shall never forget. As the 
great white leader bearing the American flag emerged from behind a boulder, 
the camels, strting out in a seemingly endless line, swung past the tents and 
broke into three even files ; then they knelt to have the loads removed, scram- 


bled to their feet again and wandered down the hill-slope to the plain, nibbling 
at the vegetation as they went. 


It is amazing how qmckly a caravan can be loaded or unloaded. Each 
box is roped in such a way as to leave a loop at the middle of the upper, inner 
edge. As the loads are lifted up simultaneously on either side, one loop is 
pushed through the other, a stick is inserted, and the weight of the boxes holds 
them in place. When vmloading, the stick is pulled out and the loads drop to 
the grovmd on either side of the kneeling camel. Very often an antelope 
horn is used in place of the stick. A hundred camels can be unloaded in fifteen 
minutes. One man handles twenty camels ordinarily, but for our caravan of 
seventy-five animals we had five Mongols. 

The best pack-saddle consists of a long strip of heavy felt woimd between 
and over the two humps and held in place by a light wooden framework. The 
Chinese often use oblong burlap bags stuffed with straw, in place of the felt, 
but they are not as satisfactory. 

The camels feed during the day on the sage and thorny desert vegetation; 
they do not like green grass and will not thrive if forced to eat it. At night 
they are tied by the nose strings, face to face on either side of a long picket 
rope, where they chew their cuds and sleep contentedly. 


Since our first night in Mongolia the weather had been delightful. During 
the day it was warm enough to be comfortable without a coat; and at night 
the temperatiire dropped to only a little below the freezing point. We had been 
fortunate in striking the period of fine weather which usually begins about the 
middle of April and continues for two or three weeks. We call it the "false 

On the afternoon of our arrival at Tuerin a tremendous "wind devil" 
swept down upon us from the north just at sunset. It came Kke a tornado, 
bringing a swirl of yellow dust and sand. We were well-nigh blinded, but even 
above the roar of the storm sounded the clatter of tins and the sharp rip of 
cloth as oiir camp was half demolished. The wind devil whirled through the 
mouth of the amphitheater and across the plain with the speed of a race- 
horse, but it left a heay>^ gale roaring among the rocks. The temperature 
dropped thirty degrees in a few minutes, and from that day until the end of 
Jtme we were seldom without oiu" sheepskin-lined coats. 




TuERiN is particularly interesting to a zoologist, for there the desert 
begins to merge into the northern grasslands and there one enters the North 
Mongolian Life Zone. The mammalian faima resembles that of the southern 
grasslands, differing only specifically in most cases, but defined by the presence 
of a few mammals which are never fotmd south of the desert. 

Just beyond Urga, one hundred and fifty mUes to the north, the North 
Mongolian Zone meets the edge of the Mongolian-Siberian Zone; this is 
sharply delineated by the forest line where it encroaches upon the rolling 
grasslands. Thus, when crossing Mongolia from the south to the north on 
the Kalgan-Urga road, we pass through, or meet, four distinct faunal zones: 
From Kalgan to P'ang Kiang, the South Mongolian Zone of the grasslands; 
from P'ang Kiang to Tuerin, the Gobi Zone of typical desert life; from Tuerin 
to Urga, the North MongoHan Zone of grasslands ; and from Urga northward 
the Mongolian-Siberian Zone characterized by many Siberian species and 
heavy forests of larch, spruce, pine, birch and maple. 


At Tuerin our traps yielded many small mammals only specifically unlike 
those of the Inner Mongolian grasslands. A five-toed, long-eared jerboa, 
Allactaga mongolica, was abtmdant; this species has a wide range in similar 
country to the west and extends northward right up to the forest margin. We 
also obtained examples of the three-toed jerboa, Dipus sowerbyi, which is a 
typical desert species and in this region reaches its northern limit at Tuerin. 
The hamster, Cricetulus griseus ohscurus, which we caught at Tuerin, extends 
along the northern edge of the desert for a long distance to the west. Another 
hamster, Cricetulus migratorius curtatus, is a semi-desert form and has been 
described as a new race by Dr. Glover M. Allen. He says: "This appears 
to be one of the Cricetulus migratorius group, the center of whose range is 
western Asia to the borders of western Europe, and is apparently another of 



the species of that area to have made its way through the Altai region into the 
eastern Gobi Desert. In addition to being apparently the most eastern mem- 
ber of the migratorius group yet discovered, it is also the shortest tailed."' 

Tuerin is the southern limit of the marmot, Marmota bobak, in this region. 
It is essentially a grassland species and I have never seen it in the desert. The 
marmots have all been exterminated in the immediate vicinity of the road, 
but their deserted burrows were abvindant not far from the telegraph station. 
They begin to appear in numbers twenty miles north of Tuerin where there are 
fewer Mongols. I will describe their interesting habits in later pages. 

Wolves and foxes 

Along the road to Tuerin we saw only one wolf and one fox. From speci- 
mens which we obtained farther to the west. Dr. G. M. Allen, 1929, has identi- 
fied the wolf as Canis lupus laniger and says that it does not seem to be very 
different from the tj'pical Etiropean race. He thinks that the name laniger, 
based on the wolf of Tibet, is also appUcable to those of Mongolia and north 
China. I cannot understand why there are comparatively so few wolves and 
foxes in Mongolia. On the prairies of western America, coyotes and foxes are 
common even to-day; in Mongolia the thousands of antelope and smaller 
mammals and birds would furnish an abundance of food, and yet one may 
travel for hundreds of miles without seeing a wolf or a fox. As a matter of 
fact, they most often appear along the main caravan trails, which are usually 
well marked by the carcasses of camels and oxen. Certainly killing by Mongols 
could have little or no appreciable effect upon the numbers of wolves and foxes, 
for the population is much too sparse and on the plains the natives seldom 
hunt them unless they have been anno>-ing the sheep herds. The Mongols 
do not use poison, as a rule, for it is too difiictilt to obtain and they are afraid 
of killing their own dogs, which are highly prized. At Tuerin I saw several 
abandoned wolf dens among the rocks. The animals seldom go far away from 
the rocky country'; the foxes have similar habits. In 1930, at Wolf Camp, 
forty miles southeast of Iren Dabasu, wolves were abtmdant, and thirteen 
were killed in the immediate vicinity of our tents. We saw many more wolves 
here than in any other part of Mongolia. 


The interesting sand-grouse, Syrrhaptes paradoxus, was stiU with us at 
Tuerin, although it is essentially a bird of the desert. We had seen it in great 
flocks in the most arid parts of the Gobi, about Iren Dabasu and Ude, where 

' Allen, Glover M. 1925. "Hamsters Collected by the American Museum Asiatic Expeditions," Amer. 
Mus. Noi'ilales, No. 179, pp. 3-4; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., 
I, No. 52, pp. 1-7, 1926. 


our cars frequently frightened them from their dust baths in the road. The 
great bustard, Otis tarda dybowskii, which had but recently arrived, breeds in 
great numbers on the Tuerin plain. The Mongolian larks, Melanocorypha 
mongolica, filled the air with song. This lark is a favorite cage bird of the 
Chinese, in fact, thousands of nestlings are caught yearly to be sold in the 
markets of north China. Not only does it have a charming song, but it is a 
mimic of no mean ability. Several larks that I had in Peking could imitate 
the mewing of a cat so perfectly that I was often deceived, and my police dog, 
Wolf, was kept in a state of perpetual excitement. 

Hundreds of demoiselle cranes, Anthropoides virgo, were performing their 
mating antics on the plain. It was most amusing to watch the male strutting 
about the apparently indifferent female, leaping into the air and doing a veri- 
table dance with wings half spread. These birds lay their eggs in late May or 
early June, but make no nest. 

The beautiful rock pigeon is pretty well distributed over Mongolia. It 
looks exactly like a domestic pigeon except for the white rump patch which is 
conspicuous in flight. The bird is an inhabitant of rocks and moimtain foot- 
hills and is not found in the desert. 

Of all the birds at Tuerin, the one which surprised and interested us most 
by its unusual habits was the ruddy sheldrake. There is no water, except wells, 
within many miles of Tuerin, and yet about twenty sheldrakes had taken up 
their residence among the granite rocks. All day long we could hear their 
mournful notes as they circled about camp and contested for a favorite roost- 
ing place on one of the highest peaks. Often we would see one silhouetted against 
the sky on the very siunmit of a ragged pinnacle looking more like an eagle 
than a water-bird. We found them throughout the desert. There, I suppose, 
they feed as do cranes upon grasshoppers and other insects, but I am sorry 
to say that I stirprised one pulling lustily at the decaying flesh of a defunct 
camel. Although we did not actually find their eggs, I am certain that they 
were nesting among the rocks of Tuerin, for the twenty or more birds were 
obviously in pairs. Later, in every marsh of the western Gobi we would see 
them glowing like molten gold among the green grass and reeds. 

The sheldrakes seemed to have little fear of the great golden eagles, 
Aquila chrysaetos daphnea, which soared about the Tuerin peaks. This eagle, 
as well as that which I provisionally identified as the steppe eagle, Aquila 
nipalensis nipalensis, had been present all the way across Mongolia. They 
were abundant, particularly in the grasslands, and at every camel carcass 
along the road we could be certain of seeing two or three. 

Mongolia has a great number of raptorial birds. Half a dozen species of 
hawks are common, but the black-eared kite, Milviis lineatus, is most abundant. 
We never camped, even in the most arid parts of the desert, but that many 


kites continually circled about the tents waiting for scraps. At Tuerin there 
were literally hundreds of kites nesting among the rocks. 


Just to the west of the road lies the Tuerin Monastery. My first sight of 
it years ago left me with the impression that it was the most remarkable group 
of human habitations I had ever seen. Familiarity in later years with the 
great monasteries of Mongolia has not changed that feeling. Coming upon 
them suddenly, as one usually does, from the empty vastness of the rolling 
plains, their outlines enhance the atmosphere with medieval strangeness and 
mystery. The Tuerin Monastery is by no means one of the largest, but it is 
very picturesque. Three temples lie in a bowl-shaped hollow among the granite 
outcrops, surrounded by hundreds upon hundreds of tiny narrow houses painted 
white with red trimmings. To the east rises the ragged mass of granite peaks; 
on the other three sides the "city" is encircled by huge piles of argul (dried 
dung) which had been collected by the priests or bestowed as votive offerings 
by devout travelers. Vast as the supply was, it would require many more 
tons to warm the houses of the lamas during the bitter months of winter. 
The dwellings and temples are built of sawn boards brought from the forests 
north of Urga. 


More than a thousand lamas live in the Tuerin Monastery. They have a 
well-organized system of church rule, and are (1922) directly controlled from 
Urga, which is only one hundred and fifty-two miles away. I have never studied 
lamaism and, rather than quote what I have gleaned from the ignorant priests, 
I would prefer not to discuss such a complicated religion. Suffice it to say 
that most of the lamas themselves have only the vaguest impression of what it 
is all about and are chiefly concerned with material things. As a whole they 
are a bad lot, and I believe that the decadence of the Mongol race is more 
directly attributable to the introduction of this pernicious religion than to any 
other cause. This I shall again touch upon in Chapter XII. 

It has been estimated that more than two thirds of the male population of 
Mongolia are lamas and I should say that this statement is not exaggerated. 
The Mongols are fanatically superstitious but they certainly are not deeply 
religious. Their chief concern is to pay sufficient respect to the gods and the 
forms of their religion to ensure themselves from the wrath of the deities. The 
majority of the lamas are human parasites, mentally and morally degraded, 
who exist by contributions from the lay population. To ensure such main- 
tenance they must keep superstition active in the minds of the populace. 
This is not a difficult matter with such simple, uneducated people as the Mon- 


gols. During my years in Mongolia I have often cured various forms of illness 
or injury, but the lama doctor never failed to require payment for his prayers. 
Usually the patients would admit that it was my medicine that had given them 
relief, but were convinced that the priest would call down ciu-ses on their sheep 
and ponies if they did not meet his demands. 

Since Mongolia has come tmder Soviet influence the Russians have made 
strenuous efforts to break the power of the lama church. As yet they have not 
dared to use force to that end but have confined their propaganda to speech- 
making, posters, writings and other more subtle methods of undermining 
lamaism. They have not allowed the appointment of another Hutiikhtu, or 
Living Buddha, since the death of the last one in 1923, and have virtually 
eliminated the lamas from important governmental affairs. 

I doubt that the influence of the religion can ever be eliminated from the 
life of the ordinary Mongol. He is a child of nature with a simple, unin- 
quiring mind. Many natiu-al phenomena he cannot explain except on the 
grounds of divine interference and, to his mind, this answers everj^hing satis- 
factorily. Superstition was instilled in his ancestors for centvu-ies and has 
been handed on to him. Since he has inherited his primitive instincts his 
belief in the power of the lamas will persist, although it may be forbidden by 

As usual, the interior of all the Tuerin temples was scrupulously clean. I 
have never known a dirtier people than the Mongols, but they must believe 
that "cleanliness is next to godliness" so far as their religious edifices are con- 
cerned. In the main room of the central and largest temple was a statue of the 
Buddha, draped with gay-colored silken strips; below it was an altar bearing 
ever-Ughted candles. Rows of prayer mats facing the center were arranged 
upon the floor, and on the left of the altar was a cushioned chair for the chief 
lama. The walls were adorned with paintings representing the "wheel of life" 
and various gods and goddesses — some of them lewd in the extreme. Streamers 
of red and blue silk hung from the ceiling and about the door. 

Near the temples and at other convenient spots were more than a dozen 
prayer wheels. Each wheel is a hollow cylinder of varying size, standing on 
end and embellished with Tibetan characters in gold. The wheels are some- 
times filled with thousands of slips of paper upon which is written a prayer or a 
sacred thought and each revolution adds to the store of merit in the future 

Near the Tuerin Monastery there were obos, or religious monuments. 
The obo is a conical pile of stones built on a hill or promontory or some other 
commanding point. There are thousands of them in Mongolia. Those near 
temples often have swords and spears of wood fixed on the svunmit. On 
almost every caravan trail, where it crosses any particularly high or difificult 


hill, an obo is built, and is replenished by every traveler who reaches that 
spot in safety, so that often they are of enormous size. My own camel men 
never passed an obo without tossing a stone or two upon the pile. Often hatas 
are hung on obos. The hata is most important in Mongol life. It is a strip of 
light blue silk two or three feet long and a foot wide. It is a token of friendship 
or good will. When a Mongol wishes to pay respect to an official or an ordi- 
nary individual, or to make a request, he usually first presents a hata. 


Tuerin had been the scene of a massacre of Chinese soldiers during the 
terrible winter of 1921 when Baron Ungem-Stemberg assisted the Mongols to 
drive out the Chinese. Everywhere we saw heaps of empty rifle shells, car- 
tridge clips and parts of uniforms. Foiu- thousand Chinese soldiers had been 
almost annihilated by three hundred Mongols imder the lead of a famous 
Mongol general whom I met several times in Urga. 

Baron Ungem had sent Cossacks to attack the Chinese, but the Mongol 
general wanted the fun himself. By doing miles of hard riding across the plains 
they reached Tuerin at daylight, before the Russians arrived. With only a 
few minutes to rest their ponies they attacked; the general himself told me 
the story of the fight. "We rode at full speed through the camp," said he, 
"killing every one we saw. Then we rode back again. The Chinese ran like 
sheep and we butchered them by hundreds." After the first few rushes the 
Mongols did not waste their ammunition, but clubbed the frenzied Chinese 
with their rifles or sabered them as they ran. Those who escaped the first 
onslaught fled into the desert half clothed. The Mongols hunted most of them 
down and virtually all whom they missed froze to death within a few hours. 
The temperature was forty degrees below zero. (See Chapter XII.) 

As I heard the story from the barbaric-looking general, sitting between 
two attendants with peacock plimies streaming from their hats, I thought of 
how like the ancient Mongol warfare was this episode of the twentieth cen- 
tury. Except for the modem weapons, the tale might have been a thousand 
years old. Had not wild hordes of Mongols, imhampered by a commissary, 
and making almost superhuman marches, conquered all of Asia and half of 
Europe in the twelfth centur>^? Then as now, if there were no mare's milk 
or meat they tightened their belts and laughed at hvmger. Thirst was their 
daily portion; he was not a man who coidd not go thirty-six hoiu-s without 
water and fight at the end. It was their heritage from Genghis Khan, an 
awakening of the old fighting spirit which is only dormant in the Mongol of 

It was this same Mongol general who rode with ten followers from Urga to 
the stronghold of the False Lama in the Black Gobi in 1923. This almost 


These graves exist in many places in tlie grasslands of Outer Mongolia. 



Tile remains of a skirt composed of bits of fossil shells may be seen about the middle of the skeleton. 



Fortunately this barbarous method of punishment is no longer used in Uiga. 


mythical figure had established a small kingdom of his own and ruled independ- 
ently of any authority. The general rode openly into the fortress, demanded 
to see the False Lama and shot him with an automatic pistol. By this cour- 
ageous act the ustirper's power was broken and his adherents scattered. 


My plan was to go to Urga to conclude the diplomatic arrangements and 
give Shackelford an opportunity to photograph a great festival which was to 
take place on May 9. The caravan was to wait at Tuerin until it was joined 
by Granger and the geologists who had remained at Iren Dabasu; I was to 
send them instructions where to join us. 

Following the storm on the evening of April 28, the day of oiir arrival at 
Tuerin, the temperature hovered about the freezing mark even during the 
day. Although the sun shone in a cloudless sky, a bitter wind continually 
swept across the plain. On May i , black clouds hung about the northern hori- 
zon and there was every indication of a blizzard, but we had no snow at 
Tuerin. The next morning dawned clear and warm, and we started for Urga 
in one heavily loaded car. 

For thirty rmles north of Tuerin the great plain spreads out on every side 
like a gently billowing sea. At first it seems as flat as a table, for the swells 
merge indistinguishably into a level whole. It is only when some approaching 
object dips for a little out of sight as a depression swallows it up that one 
realizes the unevenness of the land. 

Seventy miles beyond Tuerin we stopped at a Mongol ytirt for tea. In the 
distance snow-capped hills shone against the sky, and a dazzhng white blanket 
lay over all the land. The snow was four inches deep, and soft and slushy. It 
was well-nigh impossible to keep to the road and soon we were badly mired. 
Unloading and hard digging became the order of the next two hoiu-s. When 
the car was clear we did not go a hundred yards before it was up to the hubs 
in another mud hole. Fortvmately the snow lay in a narrow strip only ten 
miles wide, and beyond it the road stretched out brown and hard. 


We came then to the rolling grassy hills and never left them until we 
reached Urga and the Tola River. These are the true northern grasslands. 
Now they were bare and brown, but when I first saw them in early August 
they were luxuriant with long sweet grass. Then, horses and cattle grouped 
themselves into moving patches and fat-tailed sheep dotted the hills like drifts 
of snow. Great masses of bluebells covered the valleys with an azure blanket, 
and all the hills were dotted with briUiant bits of color from myriads of 



Twenty miles from Urga we saw the low mountain range which bounds 
the southern side of the Tola valley. Coming as we did from the south, we 
could see, on the mountain tops, a ragged skyline of sentinel larches made by 
the edge of the forest which covers almost every slope on the northern side. 
The moisture-laden winds sweep down from the Siberian steppes and precipi- 
tate their rain on the northern slopes in veritable deluges, all through the stun- 
mer months. There must be, however, some other reason, not so obvious, why 
even in the midst of the forest area most of the hills are bare on the southern 

The mountain facing Urga is known as the Bogdo Ola and is a very sacred 
spot. This mountain with its wonderful forest of spruce, larch, pine, birch 
and maple was a preser\'e for many kinds of game. Roe deer, wapiti, moose, 
boar, and musk deer foimd a safe refuge, for the mountain was patrolled by 
hundreds of lamas and it was worth one's life to fire a gvm or kill any animal or 
bird within the sacred precincts. For a foreigner even to go into the forests a 
permit was necessary in the days of the Hutukhtu. 


The road to Urga descends a mile-long slope to the rock-strewn valley of 
a stream tributary to the Tola River, which empties into Lake Baikal. In the 
old days there were half a dozen Chinese huts at the ford. These might be 
called the "suburbs," for the Sacred City is still a score of miles away. The 
last twenty miles was very bad going for the car. Mud, stones and ruts make 
some parts well-nigh impassable if there has been a little rain, and it is difficult 
to leave the road because of the river on one side and steep slopes on the other. 

We crossed the Tola by what was known as the "Russian Bridge." The 
river is less than a himdred yards wide at this point and flows swiftly between 
low banks bordered by willows. It has an abundance of fish, many of which are 
the typical Lake Baikal species. Huge lake trout, four or five feet long, and 
many other fish are taken in nets and sent to Urga for sale to the Chinese and 
other foreigners. The Mongols do not eat fish. 

After crossing the river we turned westward down the Tola valley and 
passed the barracks which in the old days were occupied by the Russian guard 
and later by Chinese troops. Thej^ stood deserted and partly wrecked, but 
they have since 1922 been rebuilt and occupied by Soviet soldiers. 



The city of Urga is strung out for five miles along the north bank of the 
Tola River. One comes first to the Chinese quarter, Mai-mai-cheng, then to 
the Russian section and finally to the city proper. It is not the same Urga 
that I first saw in 191 8, but the change is more internal than external, and I 
cannot better the description I wrote in 1921 : 

"Far up in northern Mongolia, where the forests stretch in an unbroken 
line to the Siberian frontier, lies Urga, the Sacred City of the Living Buddha. 
The world has other sacred cities, but none like this. It is a relic of medieval 
times overlaid with a veneer of twentieth-century civilization ; a city of violent 
contrasts and glaring anachronisms. Motor cars pass camel caravans fresh 
from the vast, lone spaces of the Gobi Desert; holy lamas, in robes of flaming 
red or brilliant yellow, walk side by side with black-gowned priests; and 
swarthy Mongol women, in the fantastic headdress of their race, stare won- 
deringly at the latest fashions of their Russian sisters. 

"We came to Urga from the south. All day we had been riding over roll- 
ing, treeless uplands, and late in the afternoon we halted on the svunmit of a 
hill overlooking the Tola River. Fifteen miles away lay Urga, asleep in the 
darkening shadow of the Bogdo Ola (God's Movmtain). An hour later the road 
led us to otu: first siurprise in Mai-mai-cheng, the Chinese quarter of the city. 
Years of wandering in the strange comers of the world had left us totally 
unprepared for what we saw. It seemed that here in Mongolia we had dis- 
covered an American frontier outpost of the Indian fighting days. Every house 
and shop was protected by high stockades of unpeeled timbers, and there was 
hardly a trace of Oriental architecture save where a temple roof gleamed above 
the palisades. 

"Before we were able to adjust our mental perspective we had passed 
from colonial America into a hamlet of modem Russia. Gayly painted cot- 
tages lined the road, and, imconsciously, I looked for a white church with 
gilded cupolas. The church was not in sight, but its place was taken by a huge 
red building of surpassing ugliness, the Russian Consulate. . . . 

"For two miles the road is bordered by Russian cottages; then it 



debouches into a wide square which loses its distinctive character and becomes 
an indescribable mixtvire of Russia, Mongolia and China. Palisaded compounds 
gay with fluttering prayer flags, ornate houses, felt-covered joirts, and Chinese 
shops mingle in a dizzying chaos of conflicting civilizations. Three great races 
have met in Urga and each carries on, in this far comer of Mongolia, its own 
customs and way of life. The Mongol yurt has remained tmchanged; the 
Chinese shop, with its wooden counter and blue-gowned inmates, is pure Chi- 
nese; and the ornate cottages proclaim themselves to be only Russian."' 


When I visited Urga in 1919 it was still in the days of Mongol freedom; 
one came and went as freely as though one were on the open plains. Before 
1922, all that had changed. Entering Urga in 1922 and in 1925 was an ordeal 
and leaving it was worse. At the outermost limits of Mai-mai-cheng, Buriat 
soldiers piled upon the car and for hours one had to submit to an inspection of 
passports, searching of baggage and person, questioning by Secret Police and a 
dozen other annoyances. In short, one was treated as a spy and a generally 
undesirable character. With evident reluctance the authorities may admit that 
at first sight there is no reason to detain the traveler, but from the moment he 
enters Urga until he leaves, one or more spies are detailed to watch and report 
upon his every movement. Woe to him who has not destroyed every scrap of 
written or printed matter before he falls into the clutches of the Secret Police ; 
even carrying oiu" best American magazines may land him in jail for a night, until 
the authorities have perused them in the hope of finding "seditious literature." 

In the old days almost every race of Central Asia might be seen on the 
streets of Urga. It was a Mecca for the pilgrimages of devout lamaists, and 
in its kaleidoscopic mass of life and color the city was like a great pageant on 
the stage of a theater, with the added fascination of reality. But now there 
are not so many dashing horsemen on the street ; not so many strange costimies 
and half-dazed nomads from the steppes of Tibet or the deserts of Turkestan. 
Their places have been taken by Russians and swaggering Buriats, and the 
great open square is daily filled with squads of awkward Mongols being drilled 
as soldiers. Machine guns stand where lines of camels knelt before and the 
prison is filled to overflowing. Even the name of the city has been changed. 
Officially "Urga" is no more — now it is "Ulan Bator," the Red City! 

Most of this happened after 1922, for at that time the Russian-Buriat 
influence was just beginning to be strongly felt. The Buriats are a northern 
tribe of Mongols who live within the borders of Siberia and are Russian sub- 
jects. Most of them speak Russian and have been brought up with Russian 

' Andrews, Roy Chapman. 1921. "Across Mongolian Plains." D. Appleton & Co., New York. 
Pp. 62-64. 


customs and ideas. The obvious plan of the Soviet is to make MongoHa a 
Bixriat country and thus far they have been remarkably successful. 

By 1925, Soviet domination in Outer Mongolia had become so complete 
that the foiu" reigning khans were virtually stripped of their power. One or 
two of them had "voliuitarily" relinquished all their land and princely claims 
and reduced themselves to the status of ordinary citizens; in rettim for this 
sacrifice they had received — nothing! 

The Mongols themselves like the change no better than the foreigner. 
The most dominant characteristics in the Mongol nature are independence 
and love of freedom; to be boimd by such useless restrictions in the Sacred 
City has robbed him of what he loves most in life. As a result many of the 
pure-blood Mongols have gone to the open plains and Urga has become a 
Buriat city vmder Soviet domination. 


Our entry into Urga was facilitated by the good offices of Messrs. Larsen 
and Hansen of Anderson, Meyer & Company, who persuaded the officials to 
postpone the inevitable questioning and examination tmtil the next day. 
Then began a fortnight's negotiations for permission to continue our work, 
which nearly shattered my nerves. It is needless to detail the process, but 
suffice it to say that had it not been for Mr. T. Badmajapoff, Adviser to the 
Minister of Justice, and for Mr. F. A. Larsen, we never would have been able 
to satisfy the government requirements. 

The officials seemed to feel that it was not reasonable to suppose that 
such a great expedition with so many camels and motor cars, and obviously 
costing such a large sum of money, could be coming to Mongolia merely for 
scientific work; that we must have some vdterior purpose in view, such as 
searching for oil or gold, or even spying for the Chinese government. 

Colgate had to drive to Tuerin with two Mongols and a Secret Service 
agent to "inspect" the caravan. He did the three-hiindred-mile trip in a day 
and a half, and on his return reported all well in camp. 

On the same day. May 6, the American Consul arrived from Kalgan with 
a letter from Granger, whom he had seen at Iren Dabasu. It brought the first 
news of the discoveries which I have already detailed in Chapter IV. It may 
be recalled that I had left for Tuerin the day after our arrival at Iren Dabasu. 
Granger's letter is of such historic interest that I quote it in full: 

"Iren Dabasu, May 3, 1922. 
"Dear Roy: 

"We're still here but expect to leave for Tuerin by the 6th. 

"This has been an unusually interesting spot, i) Cretaceous with dino- 


saurs. 2) Eocene or Oligocene with a small mammalian fauna including a 
beast very much like Titanotherium. 3) Miocene or later with a small and very 
fragmentary collection of remains, but including a remarkable beast, either 
the same as or closely related to Forster Cooper's Baluchitherium, one of the 
largest land mammals. 

"Berkey is writing an announcement for Science and a letter to Professor 
Osbom, both of which will be held until we see you. This is the first Cretaceous 
to be noted in Eastern Asia and is an important discovery. 

"Everything is all right in camp. We may have a time getting all our 
equipment and the four Chinese in the cars but will manage it somehow. 
Expect to take about three days to Tuerin. 

"Mr. Sokobin is in to-night and takes this to Urga. 

"Hastily yours, 

"Walter Granger." 

This splendid news from the party at Iren Dabasu renewed my patience 
in trying to convince the officials that we were in Mongolia to do only scientific 
exploration and had no interest whatever in politics or the commercial aspects 
of the country. 


During the time that the diplomatic negotiations were going on we had 
an exceptional opportunity to see the great festival of the Maidari which took 
place on May 9, 1922. The Maidari, or Coming Buddha, is a most sacred 
Bodhisattva. A gilded image of him reposes in a splendid temple in Urga. 
On this day, which is kept in honor of his incarnation, his image is placed on a 
huge throne, smothered in decorations and drawn through the streets as the 
central figure in an elaborate procession. 

The festival began in the early morning, for the Ivlaidari had a long way 
to go. At ten o'clock, when we reached the main square, the procession had 
not yet appeared, but the air was throbbing to the boom of drums and the 
deep notes of conch-shells. As the waves of soiond beat down upon us, we could 
see in the east a great mass of color, advancing slowly. Soon groups could be 
distinguished; then slender lines and huge -umbrellas blazing in the sunlight. 
Every shade of the spectrum was repeated a hundred times in the gorgeous 
pageant of marching lamas. As the procession neared us, I recognized the 
Premier in a robe spun of gold with a priceless sable hat upon his head. Beside 
him were the fovtr reigning khans, or kings, of Mongolia, and behind them a 
double row of princes, dukes and lesser nobles dressed in dark blue gowns with 
brilliant cuffs and streaming peacock plumes. 

The great throne bearing the Maidari was shaded by a silk vmibrella of 


rainbow colors and surrounded by the highest lamas, resplendent in cloth of 
gold. From the throne silken ropes led off to flanking lines of red- and yellow- 
clad lamas, bearing huge umbrellas of bright-hued silk. Behind t»he Maidari 
came other lamas, thousands of them, and women dressed in rich gowns with 
ropes of pearls about their necks and hair ornaments of gold studded with 
precious stones. Almost ten thousand lamas "were with the Maidari, and two or 
three thousand men and women and children followed. When the procession 
reached an open square, overlooked by the great temple on the summit of the 
hill, the throng was halted and the lamas seated themselves upon prayer-mats 
in converging masses of solid color, with the Premier in the center and the high- 
est lamas flanking the Maidari. 

The seated priests were given tea and food while a red-robed lama in the 
Maidari's chariot energetically thumped the heads of the populace with a long 
stick padded at the end. There could not be the slightest doubt in the mind of 
the suppliant that he had been blessed after the ball at the end of the stick 
landed on his head, for at times the officiating lama took huge delight in bring- 
ing it down with force enough to rock his victim. Nevertheless, thousands of 
people crowded about the throne and the priest laid on lustily for an hour. 

The splendor of the princesses and wives of the higher nobles made one 
gasp for breath. The wife of one of the great khans in particular was the most 
magnificently adorned creattu-e I have ever seen. According to the custom of 
the northern Mongol women, she had her hair plaited over a frame into two 
enormous flat braids, curved like the horns of a mountain sheep and reinforced 
with bars of gold. Each horn ended in a gold plaque, studded with precious 
stones, and supporting a pendant braid like a riding-quirt; this was enclosed 
in a long cylinder of gold, heavily jeweled. On her head, between the "horns," 
the lady wore a gold filigree cap studded with rubies, emeralds and tiu-quoises, 
and surmounting this, a ' 'saucer' ' hat of black and yellow, richly trimmed with 
sable. Just above her ears great ropes of pearls hung from her gold cap half- 
way to her waist. Her skirt and jacket were of rich silk; over aU was thrown 
a dazzling brocade coat with prominent piiffs upon the shoulders. 

The princesses had a dignity that was very becoming to their high estate. 
They were accorded none of their husbands' privileges, so far as the proces- 
sion was concerned, but, each accompanied by a servant, moved majestically 
in the midst of the vast crowd. Now and then they stopped to talk quietly 
for a moment with a friend or to acknowledge the deep salutes from both men 
and women by a slight bend of the head and just the ghost of a smile. 


Urga is in the region of the Khalka tribe of which Genghis Khan was a 
member and which, since his time, has been the dominant tribe in the north. 


The dress and hair adornment, which I have described in speaking of the 
princesses at the Maidari festival, is true in a lesser degree for all these north- 
em women. In ever>'day life the long hair braids are not enclosed in the cylin- 
ders of gold and silver, and the ropes of pearls or beads are used only when in 
full dress. The silver filigree cap and the hat of black and yellow are always 
worn. The skirt and jacket are usually replaced with a less ornate gown, but 
this is always of brilliant color; in fact, color is just as characteristic of the 
Mongol women as it is of the Burmese. 

The men are hardly less brilliant than are the women, but their dress is 
more simple in design. The lamas have robes of either flaming red or yellow, 
cuffed with blue. Beneath the gown, which is gathered at the waist by a 
twisted sash, are loose trousers. On their feet they wear great boots with 
pointed, tumed-up toes and flat soles, always many sizes too large for the 
wearer, but padded with socks of wool or fur as the weather grows colder. The 
boots are designed for warmth and comfort in the saddle, for a Mongol never 
walks if he can possibly avoid it. 

On the head both men and women wear a hat which looks like a saucer. 
It has upturned edges of black velvet and a narrow cone-shaped crown of 
brilliant yellow. Two broad red streamers are usually fastened to the rim at 
the back, or, if the wearer be an attendant to a nobleman, a short pliune of 
peacock feathers to the peak of the crown. 

The lamas, of course, always shave the head. The lay Mongols are called 
"black men" because the hair grows long and is twisted into a queue. The 
black men usually wear a gown of plum color with sky-blue cuffs. 


Urga is so filled with interesting things and personaUties that it wovdd 
reqviire several chapters to give even a sketch of them all. We explored the 
city pretty thoroughly during the time that we were waiting for the diplomatic 
negotiations to be concluded and Shackelford exposed many thousands of feet 
of motion-picture film. When we left he carried with him a record of Urga 
life such as never can be duplicated, because already much of it has disappeared, 
due to changes in the political situation. 

In 1922, the Living Buddha was in very bad health; he died the following 
year. He was blind and few people except his most intimate advisers saw him. 
I called to present a Savage 250-3000 rifle, which was gratefully received by 
one of the high lamas who doubtless disposed of it immediately for a good price. 
In return I was given a photograph of the Hutukhtu and his wife. 

At that time his quarters were two hideous foreign houses of Russian 
style, painted a bright orange. He changed his dweUing place frequently 
and used to be particularly fond of a green-roofed collection of temple-like 




o ij 



This man was nearly eiKhl feel tall and weighed 
307 pounds. 

The Expedition made the first motion pictures of this festival. It has now been aljolished. 


buildings surrounded by a palisade painted white and green. This stood 
beside the bank of the Tola River amid a beautiful grove of birch trees just 
at the foot of the Bogdo Ola. Another larger palace stood on the flat meadow 
half-way between the river and Urga. It was of Chinese architecture and 
was fronted by a huge and extremely ornate gate. 

The garage of the palace interested me immensely because in it was a 
car, purchased by the Hutukhtu, that had made the first trip from Kalgan to 
Urga. In the garage stood a huge Fiat truck with an enclosed body capable 
of carrying twenty people. This had recently been purchased for eighty-five 
hundred dollars but probably never would be used. There were two Franklin 
cars and the old Ford relic which was the historic car to make the first crossing 
of the desert. Several other automobiles of the vintage of twenty years ago 
reposed in one comer of the building; one was a French three-wheeled affair 
with a carbureter two feet square projecting far out on the left side. How 
the machine ever reachied Urga is a mystery. Several other relics were too 
broken and covered with dust even to be identified. The only really workable 
thing in the entire garage was a motorcycle. 

A short distance from the great palace in the meadow were the remains 
of the Living Buddha's zoological garden. At one time I believe he had quite a 
collection, chiefly of Mongolian animals, but all had died except an elephant 
and a wapiti stag. 


The last of my many meetings with the Cabinet and officials took place on 
May 17, 1922, at nine o'clock in the morning. Then I signed, with the Premier 
and Minister of Foreign Affairs, the agreement over which we had bargained 
for a fortnight. Passports already had been obtained for every member of the 
party, for the cars, chauffeurs, gims, rifles and revolvers. This had required 
days of work and the expenditure of more than a thousand dollars. But at last 
it was ended and the final political barriers to our journey to the desert had been 
removed. We were to be allowed to carry on our expedition and in return we 
guaranteed to fulfil certain conditions. Among these was the obligation to 
take a government official with us to see that we carried out otir agreement. 
The official designated to go was our friend, Mr. Badmajapoff, with whom it 
had all been arranged previously. Badmajapoff is a Buriat who has had con- 
siderable experience with the great explorer, P. K. Kozloff . He is a master of 
Oriental diplomacy and moreover is a charming companion. His presence on 
the Expedition was most agreeable to us all and made for us a firm friend, who 
in later years did much to help me steer a safe path for the Expeditions among 
the political rocks which barred the way to the great open spaces of the Gobi 


Mr. F. A. Larsen was to accompany us as interpreter. Mr. Larsen had 
been in Mongolia nearly forty years, and had taken a most important part in 
the affairs of the country. The Living Buddha had often sent him as an 
emissar>' to Peking, when misunderstandings or disturbances threatened the 
political peace of Mongolia. Very few of the important Mongols were not 
numbered among his friends, and the nomad dwellers on the plains knew him 
by reputation. To the foreigners over all China he is familiar as the principal 
dealer from whom race ponies are obtained, and there is little about the 
Mongol pony which Larsen does not know. Moreover, he is an enthusiastic 
sportsman. His knowledge of Mongol psychology and of the country was of 
much service to the Expedition. 


At the Cabinet meeting the Premier asked that I should capture for the 
Mongolian Government a specimen of the Allergorhai horhai. This is probably 
an entirely mythical animal, but it may have some little basis in fact, for every 
northern Mongol firmly believes in it and will give essentially the same descrip- 
tion. It is said to be about two feet long, the body shaped like a sausage, and 
to have no head or legs ; it is so poisonous that even to touch it means instant 
death. It is reported to live in the most arid, sandy regions of the western 
Gobi. What reptile can have furnished the basis for the description is a 
mystery ! 

I have never yet found a Mongol who was willing to admit that he had 
actually seen it himself, although dozens say they know men who have. More- 
over, whenever we went to a region which was said to be a favorite habitat of 
the beast, the Mongols at that particular spot said that it could be found in 
abundance a few miles away. Were not the belief in its existence so firm and 
general, I would dismiss it as a myth. I report it here with the hope that 
future explorers of the Gobi may have better success than we had in running 
to earth the Allergorhai horhai. 


During the time that we were in Urga in May, 1922, conditions were 
most unsettled. It was the aftermath of the bloody days which had just pre- 
ceded our coming, when Baron Ungem-Stemberg had slaughtered Jews and 
Bolsheviks by the hundreds. 

After the Red Russians had occupied the coimtry, a systematic clearing 
out of all Mongols opposed to their policies was begun. It was a reign of terror. 
No one knew how soon he would be accused of some crime and thrown into the 
horrible prison. One might even be there for months without knowing in 
what way he was supposed to have transgressed the law. The Minister of 


War had been executed a few days before we arrived. The former Minister of 
the Interior, an old friend of mine, was vmder suspicion. I called upon him 
one afternoon. The next morning I went again on business and found his 
house in chaos. That night he had been dragged out and shot. It was not 
Mongols alone who suffered. Individuals of half a dozen nationalities, includ- 
ing Americans, were in difficulties. Murder and sudden death stalked ahead 
upon the streets. It was an exceedingly good place to leave. 

It was also sickening to see the last remnants of medieval state ruth- 
lessly destroyed in a coimtry which was normally happy tmder the rule of its 
hereditary princes and a "people's reign" of terror and unhappiness substi- 
tuted in its stead. It was like destroying a beautiful painting by an old 
master — a painting in which the colors have been only softened and enriched 
by time — and putting in its place a glaring chromo! 



On May lo, I sent one of Mr. Larsen's Mongols to Tuerin to gmde our 
cars to Bolkuk Gol, nineteen miles from Urga, where we were to join them. 
He went in a motor driven by a German mechanic, with two Chinese passen- 
gers. A friend of mine, Mr. Brandhauer, drove another car in which my wife 
and Doctor Black rode. The German's motor was rolling along peacefully on 
the great plain thirty miles from Tuerin when a wheel broke and the car cap- 
sized. One of the Chinese passengers was killed and the Mongol guide sus- 
tained a broken collar-bone and a fractured skull. Nevertheless, the plucky old 
fellow insisted upon fulfilling his duty of guiding our caravan to the rendezvous. 

The IVIongols at the telegraph station would not allow the dead Chinese 
to remain in the place. Doctor Black at last decided to bury him some dis- 
tance away. The relatives of the deceased caused Brandhauer an enormous 
amount of trouble. Eventually he had to bring up a coffin from Kalgan, 
exhume the body and return it to China, besides paying a heavy indemnity. 
Spare parts were taken from our caravan at Tuerin and the broken car was 
repaired so that it could return to Urga. 

At this time we heard rumors that the expected clash in China between 
Chang Tso-lin and Wu Pei-fu had taken place and that Chang had been 
defeated. The fact that there was no communication by telegraph and no 
mail service to Urga made it ver>^ difficult to obtain accurate news. We were 
all hopeful that Wu had been victorious, for at that time we foolishly thought 
that he might bring comparative peace to China. 

On May i8, we left Urga with Larsen and Badmajapoff, after paying a 
final call on the Premier to say "goodbye." He was a high lama, an old friend 
of Larsen's, and had stoutly supported our cause in the official councils. When 
I thanked him he said: "Yes, it has been difficult. Things are not as they used 
to be a few years ago." 


We drove over the bridge across the Tola, just below the Living Buddha's 
palace, and up the hill to the beautifiil meadow which sweeps from the base of 



the Bogdo Ola down to the willow-bordered margin of the river. Marmots 
were popping in and out of their holes, whistling cheerily, and larks flooded 
the air with song. We were all very happy, for months of work in a new coiin- 
try lay before us and the political barriers had all been lifted. 

The air was cold even in the sun. Two days before there had been a 
freezing wind and the summits of the Bogdo Ola glistened with snow; white 
patches covered the grass even on the lower slopes. 

Motor trouble delayed us five miles from Urga. I could not diagnose it 
correctly, and Mr. K. P. Albertson, who had accompanied us to see the Expe- 
dition start, went ahead to bring assistance. In less than half an hour he 
returned with Colgate, Granger, Berkey and Morris. They had been on a 
geological trip along the river and Albertson met them only a few miles away. 
Colgate soon had the motor working and in the meantime we kept up a rapid- 
fire conversation. I got the first verbal reports of the great discoveries made 
in the vicinity of Iren Dabasu, and gave them an account of my own diplo- 
matic experiences. 

As we proceeded to camp, we drove up and down grassy hills. At Bolkuk 
Gol the blue tents were pitched on a gentle slope with a great snowbank 
glistening in the distance. The men of our Expedition had had a cold time of 
it, since they had arrived just in time to meet a real blizzard, even though it 
was mid-May. Two hours after we reached camp, Merin galloped in on his 
great white camel. The caravan was close behind him. Soon we saw the long 
line of camels silhouetted on the summit of a hill, with the American flag stream- 
ing above the leader. Thus for the first time the entire Expedition was together. 

I worked imtil after dark selecting food, gas and other equipment to take 
with us in the cars, for we estimated that it would be nearly a fortnight before 
we would have contact with the camels again. Our destination was the prov- 
ince of Sain Noin Khan, about three hundred miles southwest of Urga. From 
Mongol reports, I judged it would be a fruitful field for our work. It was a 
leap in the dark, but if the country did not give us results we could always 
turn southward into the Gobi. Sending the caravan to that region would put 
it in a central position. Since it would follow our trail we could always send 
back to it for additional supplies or to give Merin further instructions, if 


When we left Bolkuk Gol on May 19, 1922, we followed the valley of the 
Tola River west by south. I had traversed this same trail in 19 19, and at the 
end of six hours' run we camped at the identical spot where my tents had been 
pitched after eight days of laborious cart travel. It was a beautiful valley. 
Wide flat meadows luxiuiant with grass bordered the river and we were seldom 


beyond a sight of a yurt or of herds of fat-tailed sheep, cattle and ponies. I 
did not try to do any zoological collecting, for my 191 9 work had shown that 
the valley had been virtually denuded of animal life by the Mongols. We 
did not see even a marmot or a hare, and not a sign of antelope. 

Geologically the valley consists largely of ancient graywacke rocks with 
interbedded shale or slate in some places. Our geologists designated this the 
Khangai Graywacke Series and found that it forms the major part of the 
mountains of the Arctic Divide, as well as the Khangai mountain range, 
through the province of Sain Noin Khan. 

Our first camp was at the point where the Tola River comes down from 
the northeast and makes a sharp bend westward. On the northern bank stands 
a large temple, but we did not cross the river to inspect it, as we had no horses. 
Dozens of lamas, however, came over to us and examined our cars with the 
greatest curiosity and fear. They circled doubtfully about the tents and it 
was not untU they realized that the strange things were temporarily harmless 
that they came in for a closer view. Then a blast of the horn or a sudden 
starting of the engine would send them scurrying in all directions with fright- 
ened shrieks. They told us that these were the first automobiles they had ever 


I had contracted a severe infection in my eyes while in Urga, and the glare 
and wind gave me such excruciating pain that I was happy when darkness 
came. As a matter of fact my eyes never recovered from the severe strain and 
it means the wearing of glasses for the rest of my life. The Mongols suffer 
from desert blindness, and eye-trouble is perhaps their most common afflic- 
tion. To soften the glare they often use pieces of wood with narrow longi- 
tudinal slits; sometimes they weave "glasses" of black horsehair, or blacken 
the face about the eyes with soot. In later years I foimd that the Crookes 
lenses were by far the most satisfactory for desert use, but the damage had 
already been done so far as I was concerned. 


After leaving the Tola River we turned directly southward and struggled 
through ten miles of what we called "nigger heads." The geologists, however, 
more properly termed them playa tufts. They are roimded mounds, which 
have been formed in sandy country' by wind about the base of desert vegeta- 
tion or some other obstruction. They vary from one foot to six or eight feet 
in height and usually are held together by the roots of low thorny bushes. 
In a low damp region the "nigger heads" are small, rotmded, grass-covered 
hummocks such as are common in the marshes of America. From a distance 


they give the terrain the appearance of the greatly magnified pebbled surface 
of a golf ball. Although their mode of origin differed according to the particu- 
lar region in which they occurred, the trouble they gave our cars was a constant 
factor. There was seldom a day's run in which we did not encounter a few miles 
of this distressing terrain and in a single day we have crossed as much as fifty 
miles of it. In the fields of very high "nigger heads," we could always find a 
way through them by constant twisting and turning, but we had to go over 
the smaller ones, which were usually close together. This necessitated careful 
driving, and the strain on the chassis of the heavily loaded cars was enormous; 
nevertheless, we never had any breakage from this sort of terrain. A car with 
a rigid chassis simply could not go over such country. As soon as possible 
after such an experience, all the cars were given a thorough inspection and the 
bolts and nuts tightened ; it was only by such care that accidents were avoided. 


After leaving the field of "nigger heads," we came to a low mountain chain 
of granite much like that at Tuerin ; for some distance intrusive granites in the 
form of great dikes and bosses stood out prominently among the rovmded 
grass-covered hills. The geologists felt certain that we were traveling over 
pre-Cambrian rocks, but the relationships were so obscure that they needed 
some time in which to study the region ; therefore they asked that we camp at 
a well twenty-five miles from the Tola River. I was not loath to stop, for it 
was a beautiful place and I was glad to continue my mammal collecting line 
from the spot where I had ended at the bend of the Tola River in 1919. It was 
so early in the spring that the grass was still short and no Mongols had come 
in to take advantage of what would be wonderful grazing later in the year. 

We camped early in the afternoon in a great amphitheater formed by 
roionded hills. Not far away was a well of delicious cold water. We had seen 
one wolf and thousands of antelope, Procapra gutturosa, in the wide valleys, 
and marmots were whistling and popping in and out of their holes as though 
manipiilated by strings. Granger and I, with the taxidermists, set a line of one 
hundred and fifty traps and looked forward to a new small mammalian fauna 
in the morning. 

Badmajapoff, Larsen and Colgate himted antelope and brought in five 
after a two-hour run. We christened the spot "Five Antelope Camp." 
Granger and I had a most interesting and profitable competition as to who 
could catch the greatest number of species of small mammals in a given number 
of traps. 

All of us had a busy time at this camp. The geologists were out from 
daylight until dark driving over the surrounding country in their car. Even 
though the rocks appeared to be pre-Cambrian and unfossiliferous, still they 


must work out the structure of the region and its relationship to that which 
they had ab-eady crossed. 


Badmajapoff shot a magnificent eagle owl, Bubo bubo kiantschensis, 
among the rocks at Five Antelope Camp. A few miles farther on, Larsen 
shot one that was sitting on its nest in a tiny cave in the precipice. He reported 
that there were three downy yoimg and I went up immediately to photograph 
the nest and collect the young and materials for a museum group. When only 
a short distance from the spot we saw a large dark hawk fly off with a young 
owl in its claws. One baby remained, the other two having been taken in the 
hour which had elapsed since the mother was shot. The nest was in a niche 
in the rock, four feet high, two feet wide and two feet deep. It was unlined but 
contained many feathers of birds, small bones and pellets. The pellets were 
cylindrical and four inches long by two inches thick. In them I identified a 
great number of Allactaga bones, a few Ochotona and Mus and a musteline of 
some sort. 

While we were at the nest, and for a long time afterward, the male owl 
soared like an aeroplane above the rocks. For two hours I tried unsuccessfully 
to get a shot. When we went to bed at eleven o'clock the owl called for an 
hour from the rocks; the note was a low, resonant who, who, who. 

There were a few spermophiles living in a small colony, but strangely 
enough these animals were nowhere abundant until we reached Sain Noin. 
The gophers seem to have a peculiar distribution in Mongolia. Often we would 
cross himdreds of miles where there was not a sign of the animals; then sud- 
denly we would discover a small isolated colony. This might consist of a very 
few individuals or even of thousands, as in southern Mongolia and at Sain 
Noin. Usually the colonies were not separated by any geographical or floral 
barrier, so far as we could perceive. 

We caught a beautiful little hamster with a median dorsal stripe, which I 
identified as Cricetiscus campbelli; I had first become acquainted with it at 
Urga in 1 91 9. A grassland vole, Microtus brandti, was very abimdant and we 
caught our first picas, Ochotona dauurica. Shackelford waged war upon the 
marmots with a .22 caliber rifle and with steel traps. Marmot skins are one of 
the principal exports of Mongolia and millions are sent out every year. 


In 1 92 1 I wrote a description of the Mongol methods of hunting marmots, 
which is as follows: 

"The marmots hibernate during the winter, and retire to their biirrows 
early in October, not to emerge imtil April. W^hen they first come out in the 


spring their ftar is bright yellow, and the animals contrast beautifully with the 
green grass. After the middle of Jime the yellow fur begins to slip off in 
patches, leaving exposed the new coat, which is exceedingly short and is mouse- 
gray in color. Then, of cotirse, the skins are useless for commercial purposes. 
As the summer progresses the iur grows longer until by September first it has 
formed a long, soft coat of rich gray -brown which is of considerable economic 
value. . . . 

"The natives always shoot the animals. When a Mongol has driven one 
into its burrow, he lies quietly beside the hole waiting for the marmot to appear. 
It may be twenty minutes or even an hoiu-, but Oriental patience takes little 
note of time. Finally a yellow head emerges and a pair of shining eyes glance 
quickly about in every direction. Of course, they see the Mongol but he looks 
only like a mound of earth, and the marmot raises itself a few inches higher. 
The hunter lies as motionless as a log of wood imtil the animal is well out of 
its burrow — then he shoots. 

"The Mongols take advantage of the marmot's curiosity in an amusing 
and even more effective way. With a dog-skin tied to his saddle the native 
rides over the plain until he reaches a marmot colony. He hobbles his pony 
at a distance of three or foiir hundred yards, gets down on his hands and knees, 
and throws the dog-skin over his shoulders. He crawls slowly toward the near- 
est animal, now and then stopping to bark and shake his head. In an instant, 
the marmot is all attention. He jumps up and down whistling and barking, 
but never venturing far from the opening of his burrow. 

"As the pseudo-dog advances there seems imminent danger that the fat 
little body will explode from curiosity and excitement. But suddenly the 
'dog' coUapses strangely and the marmot rises on the very tips of his toes to 
see what it is all about. Then there is a roar, a flash of fire and another skin is 
added to the millions which have already been sent to the seacoast from 
Outer Mongolia. 

"Mr. Mamen often spoke of an extraordinary dance which he had seen 
the marmots perform, and when Mr. and Mrs. MacCallie retvuned to Kalgan 
they saw it also. Mac said that two marmots stood erect on their hind legs, 
grasping each other with their front paws, and danced slowly about exactly as 
though they were waltzing. He agreed with Mamen that it was the most 
extraordinary and amusing thing he had ever seen an animal do. I can well 
believe it, for the marmots have many cvu-ious habits which would repay close 
study. The dance could hardly be a mating performance since McKenzie 
Young saw it in late May and by that time the young had already been 

■Andrews, Roy Chapman. 1921. "Across Mongolian Plains." D. Appleton & Co., New York. 
Pp. 101-104. 


Mr. J. H. Miller has given an interesting account of marmot htinting by 
the Mongols and Kirghiz of the western Altai Movmtains. He says : 

"As the old man and I rode along over the springy turf, the shrill whistle 
of the marmots resounded on every hand. By the autumn these jolly little 
animals have amassed such quantities of fat, preparatory' to their winter 
sleep, that they present a most comical appearance. Their short legs are 
almost invisible, and, as they make for their holes, they look just like large 
muffs rolling down the hillside. Though these animals, along with the snow- 
cock, add greatly to the picturesqueness of the scene, they are no friends of the 
himter. On several occasions, whilst after sheep or ibex, my staUc has been 
spoilt either by the whistle of the marmot, whose quickness of eyesight is 
almost unequalled, or by the weird cries of a covey of snow-cock as they sailed 
out over the valley. 

"During the morning we came across a marmot hunter, a wild-looking 
figure clad in tattered sheepskins, and armed with an ancient long-barreled 
muzzle-loader, with the usual forked rest. During the summer these hunters 
wander about the hiUs, carrying nothing in the way of food except a goat- 
skin sack of 'kiuniss' and a small bag of salt, tied to their saddles. They rely 
almost entirely on marmot flesh for their meat, only occasionally killing a wild 
sheep, when an easy opportunity presents itself. 

"There are two methods of hunting the marmot adopted by these people: 
one is merely to make a low breastwork of sods within thirty yards or so of a 
well-used burrow, and lie patiently behin^i it till a beast appears; the other 
requires slightly greater skill and energy. On locating the marmot outside his 
hole the hunter advances boldly towards it at a rapid walk, carrj'ing his gun 
in his right hand, and incessantly waving a bimch of white sheep's wool 
attached to a stick, or a fox's brush, in his left. This unusual sight so excites 
the ci-uiosity of the marmot that he will often sit bolt upright at the entrance 
to his hole and allow the hunter to approach close enough for a hurried shot."^ 

Marmots give birth to from two to four young each season. By mid- 
Jime the babies are about ten inches long and covered with soft yellow-white 
fur. The little fellows are very playful, and I have often seen them rolling 
about like kittens in the grass near their burrows, while one or both of the 
parents keep watch. By August 15, the young are two-thirds grown and have 
assumed the brown winter pelage. 

The large weasel, Miistela larvata tiarata, is a very real enemy of the 
marmot, and I have frequently caught these fiery little beasts in traps set in 
marmot holes. In August, 191 8, some disease took a heavy toll of marmots in 
the Urga region. I saw many lying dead beside the road and others that were 

' Carruthers, Douglas. 1914. "Unknown Mongolia," with three chapters on sport, by J. H. Miller. 
Hutchinson & Co., London. I, pp. XVIII, 1-318; II, pp. X, 319-659- 


barely able to crawl to their burrows. This was most significant, since, from 
January to April of that year, pneiunonic plague had been raging in south- 
em Mongolia, Smyuan, Chahar, Shansi, Chihli, Shantung, Anhwei and 
Kiangsu. Sixteen thousand victims had been claimed by the disease. It has 
been suspected that marmots had a very direct relation to the series of out- 
breaks of pneumonic plague in southern Siberia, Mongolia, and Manchuria. 

Dr. Wu Lien-teh, Director of the Manchurian Plague Prevention Service, 
conducted researches in Mongolia diuing 19 12 and concluded the pneiunonic 
plague epidemics arise as a secondary manifestation of bubonic plague, but 
that the pulmonary form of the disease was not directly communicated to man 
by marmots. In other words, that the Manchurian epidemics arose as a result 
of primary bubonic infection invading the lungs in addition to other organs. 
He found that chronic plague may exist among marmots, giving rise to epi- 
demics of bubonic plague as a direct result of infection from skinning or eat- 
ing the flesh. He could not be sure whether the rat or the marmot was most 
important in distributing the bubonic disease in Manchuria. His experiments 
showed, however, that marmots are easily susceptible to the pneumonic form 
produced by their inhaling the Bacillus pest is. He foimd that the mask was the 
principal means of personal protection against pneumonic plague. 

In Manchuria and Mongolia where he worked along the line of the Chinese 
Eastern Railway, Doctor Wu said that both the Siberian and Mongol settlers 
eat the flesh of marmots and that considerable quantities are salted and 
exported to European Russia. He himself ate the flesh and foimd it to be ten- 
der and to compare favorably with rabbit meat. I never saw the Mongols 
eat marmot, but probably that was because other and better meat was easily 

Doctor Wu reported that the Chinese and Russian himters along the rail- 
way snare the marmot by arranging a wire loop over the hole. In the parts of 
Mongolia that I have visited the natives knew nothing of this method of 
trapping. He also remarks that the Mongols will not touch a sick marmot or 
one that seems unable to escape when surprised, and that consequently plague 
is but seldom reported among the Mongols. On the other hand, the Russian 
and Chinese hunters handle even sick animals carelessly. 

Most marmot burrows have two or more entrances. The earth that has 
been dug forms a flat-topped moimd covered with grass which usually is some- 
what greener than that of the surrounding plain. Thus, with a little practice 
one can distinguish the dangerous holes a considerable distance away. After 
leaving the northern grasslands and turning southward into the desert, we 
saw no more marmots until we reached the Altai Mountains. There on the 
grassy slopes at seven or eight thousand feet above sea-level the animals were 
again abundant. 



May 21 was the first time that we were able to sit comfortably in the tent 
without the warmth from an "argul" fire. "Argul" is the Mongol name for 
animal droppings, which furnish the only fuel on the plains and deserts. It 
corresponds to the term "buffalo chips" of our American west. Argul makes a 
hot fire which soon bums out and must have a draught to keep it blazing. 
When we made camp it was the duty of our three Mongol interpreters to collect 
argul for the camp as soon as the tents were up. The cooks usually carried a 
supply with them so that dinner was not delayed. The Mongols have become 
so accustomed to using argul that even when they are camped on the edge of a 
forest, they prefer to collect it rather than bring in wood. Their "stove" con- 
sists of a small skeleton cylinder made of strips of iron and open at the top and 
bottom. Three-inch spaces between the iron hoops allow for a sufficient 

Our cooks used an oven which I had developed out of a Standard Oil tin. 
A cover at the end was hinged by bits of wire so it could be dropped down and 
fastened; with other pieces of wire, flanges were made, longitudinally, which 
served as rests for the trays. One of the great advantages of this oven lay in 
the fact that when one was worn out another could be constructed in an hour, 
for we always had a sufficiency of gasoline tins. The cook would dig a narrow 
trough directed toward the wind. Then with the oven rested across this 
trench on iron tent pegs, an argul fire was built below and on top of the oven. 
Biscuits, bread, meat and even pies and cakes could be baked in this simple 
oven as well as in a kitchen range. 


On May 22,, we left Five Antelope Camp for the monastery of a Mongol 
prince, Tse Tsen, on the way to Sain Noin. Our trail continued southwest- 
ward, among grass-covered hills where great granite bosses stood out like 
ramparts against the sky. It was a beautiful grazing coimtry. In the bottoms 
of several valleys streams of sweet clear water ran between banks of emerald- 
green grass. But as yet the Mongols had not moved their yurts and the coun- 
try was deserted save for antelope and birds. 


Beside the trail we foimd a brovm pit-viper, Agkistrodon halys intermedins, 
the only species of poisonous snake that I saw in Mongolia. It is about two 
feet long with a thick body and dull brown color. It is related to our American 
copperhead. Because of the long, bitterly cold winters and excessive dryness, 
the reptilian fauna of Mongolia is exceedingly limited. We caught only one 
other snake in the desert, the non-poisonous Elaphe dione, from the edge of 









the grasslands which range into Chihli and Shansi Provinces of China. The 
viper appears to have adapted itself exceptionally well to the long periods of 
hibernation and starvation, for in certain places along the face of gravel bluffs 
it was present in numbers. In fact we had several most uncomfortable experi- 
ences with them in later years. 

The Mongols are afraid of these vipers, and the sight of one in the road 
threw our interpreters into a panic. That night they refused to sleep upon the 
grovmd and betook themselves to the cars. I found that the Mongols have the 
same method of protecting themselves from snakes that is used in the Ameri- 
can west. They stretch a camel's hair rope aroimd the camp on the ground 
and believe that the vipers will not crawl over it. This is only one of the many 
similarities in life and customs, which like conditions have developed between 
the Mongols and our own westerners. 


We had become so accustomed to the deserted hills that it was most sur- 
prising to suddenly top a long slope and see below us in a wide valley the 
lama monastery of Tsetsenwang. It lay like a miniattire city of tiny houses, 
temples and pinnacled shrines, waUed in by enormous piles of argul. The 
monastery was similar to the one I have described at Tuerin, but was con- 
siderably larger. 

To the north we could see the gaping mouths of three canyons, and before 
the lamas had time to give more than a yell of surprise the whole fleet of cars 
had passed on up the valley. I selected a camping groiind well up in one of 
the canyons where we were partially sheltered from the wind. A tiny stream 
gurgled over a bed of white stones and in the next canyon to the north an ice 
floe spread out like a vast white fan ; it wotild be many weeks before the minia- 
ture glacier melted. 


Ever since leaving the Tola River and tiuning south west ward, we had 
traveled over a floor of ancient rocks, in which graywacke predominated. The 
great intrusions of granite in the form of dikes had become more and more 
frequent as we neared Tsetsenwang. It was a region that reqtdred study by 
the geologists, and as I wished to wait here imtil the camels arrived we made 
ovirselves snug for at least a week. 

Berkey and Morris spent most of their time driving about the surround- 
ing coimtry in their car and covered several htmdred miles during our stay. 
They had begvm to realize that the great granite masses which we had encovm- 
tered ever since leaving Kalgan were definitely connected and that they were 
about to make a most important geological discovery. Regarding it they have 



written as follows: "The Third Asiatic Expedition, in the early stages of its 
traverse across Mongolia, repeatedly noted occurrences of granite. Some are 
small intrusives, whereas others cover extensive areas of undetermined bound- 
aries. They are associated with so great a variety of other rock formations 
of widely different ages, that at first there was little to suggest their possible 
unity. Because of the fact also that the granites themselves show considera- 
ble variety of minor character, it was assumed that they were essentially inde- 
pendent intrusion phenomena that might have as great age differences as the 
hosts with which they are associated. 

Hoot-Pendant of Gra/wacKe 




Granite Cut fy Numerous DiKea 

Figure 8. — Margin of a roof-pendant north of Tsetsenwang. From "Geology of Mongolia." 

"As the traverse was extended northward from Kalgan, however, a cer- 
tain similarity of rock t\'pe and of field relation was noted, suggestive of a pos- 
sible common origin. This became all the more impressive in the north and 
west, where, in many places, the rock floor for tens of miles together is made 
up wholly of granite. 

"It is the purpose of this paper to indicate some of the evidence bearing 
on the structural and genetic relations of these granites, and to support, in 
more definite terms than has hitherto been done in the publications of the 
Expedition, the conclusion that they represent a great granite bathylith." 

The authors conclude: "For this bathylith, which in dimensions seems to 
compare favorably with the greatest bathyliths thus far known in other parts 
of the world, we propose the name 'The Great Mongolian Bathylith.' "' 

' Berkey, Charles P., and Morris, Frederick K. 1924. "The Great Bathylith of Central Mongolia," 
Amer. Mus. Novitates, No. 119, pp. l-ll; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. 
Hist., I, No. 24, pp. i-ii, 1926. 



In the broad valley north of otir camp there was a marshy region which 
developed into a series of salt lakes; the largest of these was at least a half- 
mile wide. Salt thickly encrusts the marsh and gives a white margin to the 
lake. Probably it is always too saline for fish, and the Mongols said that none 
were ever found there, but great ntmibers of waterfowl were resting on the 
surface. I identified swan-geese, mallards, teal, ruddy and burrow sheldrakes 
and a dozen great white swans, probably whoopers, Cygnus cygnus. Proba- 
bly these birds were nesting in the vicinity. In fact, the canyon resounded all 
day with the cries of ruddy sheldrakes, which were continually flying up and 
down the stream and gabbling among the rocks. 

There were many red-billed choughs, Pyrrhocorax pyrrhocorax, among the 
rocks of otir canyon, and "Buckshot," one of the Chinese assistants, brought 
four young to camp. He fed them on fresh meat soaked in water. Subse- 
quently they became o-ur most amusing pets, being as tame as chickens and 
flying all about the tents. They were intensely curious. I have a photograph 
of one sitting on Granger's head while investigating his ear; a few moments 
later it even put its bill into the bowl of the pipe he was smoking. 

At night we heard the howling of wolves from up the canyon. Larsen 
found a den which showed signs of containing young. We set traps in the two 
entrances while the she-wolf sat on a crag overlooking the scene, howling dis- 
mally. She was so suspicious that she did not return for two days. At last we 
dug into the den and obtained two pups as specimens; they were about the 
size of half -grown cats. The Mongols said that occasionally lynxes were seen 
among the rocks, but we found no signs of them. In the forest, north of Urga, 
I shot a beautiful lynx, Lynx lynx isabellina, but I believe that its southern 
range is where the coniferous forest meets the plains, and that this species, at 
least, would not be found on bare hills. 

On the flat ground at the entrance to the canyon there was a vast colony 
of meadow voles, Microtiis brandti. I have never seen so many individuals of a 
single species of any mammal. To say himdreds of thousands is not over- 
stating their number. The plain was literally alive with them and one could 
not step without treading on their burrows, which were connected by tiny 
paths. A continual high-pitched chirping soimded like thousands of crickets 
londergroiuid. They appeared to be almost entirely diurnal, for oior traps 
caught many more during the day than at night and we saw them at all hours. 
Granger and I caught a dozen yoixng of varying ages in our hands. 

All other small mammals were very rare at this camp ; even such usually 
abvmdant species as hamsters and jerboas were almost non-existent. It is, I 
believe, quite certain that the Microtus were responsible for this condition, 
for their great numbers had made it difficult for other small mammals to live 


there. We first encountered this species at Five Antelope Camp and it carried 
on to the south westward continuously to Sain Noin, which was as far west as 
we went in the grasslands. It was always abundant except where there was a 
gravel terrain ; evidently it is a true grassland vole. 


Tse Tsen Wang was one of the richest and most powerful of the Mongol 
princes in 1922. He was then about fifty years old, nearly six feet tall and 
beautifvilly proportioned. The blood of a long line of high-bom nobles nins 
in his veins and shows in his finely cut features. His younger brother, who 
was in Urga, was a handsome man and fulfilled all the romantic western ideas of 
how an Oriental prince should look. He was most careful in his dress, wear- 
ing long gowns of richly brocaded silk, gorgeous sashes and hats of gold thread 
or sable. He was a good deal of a "young-man-about town," a fine shot and a 
splendid horseman. He seldom went on the street without one or two retain- 
ers, who likewise were carefully dressed and wore peacock plumes in their 
hats. All the hereditary nobles of Mongolia whom I met could easily be dis- 
tinguished, by their finer features, from the ordinary natives. Some of them 
still maintain considerable state and rigidly adhere to certain forms of inter- 
course •with strangers. 

The niling prince was a fine type of Mongol aristocrat. I had met both 
him and his younger brother in Urga and had purchased several ponies from 
them. In 1919, he had a herd of ten thousand ponies and many sheep and 
camels. In 1920 the Chinese soldiers confiscated twenty-four hundred ponies, 
and the new regime of Russian-Buriat control in Urga had treated him almost 
as severely. He still had a good many sheep and camels but was fearful of 
what would happen to them in the next few months. On the day after our 
arrival, Larsen, Badmajapoff and several others drove over to call on the 
prince. At that time his yurt was about ten miles beyond the monastery, but 
he shifts it frequently as the feed changes. They reported that he lived in con- 
siderable state and comparative cleanliness. 


The next day Tse Tsen Wang retiimed the call and spent the day in camp. 
He brought a great chunk of white cheese and a skin bag of "kumiss." The 
latter, made of fermented mare's milk, is the national drink of the Mongols. 
It has rather a sour, not unpleasant, taste and is highly intoxicating. "Kimiiss" 
may be found in the yurt of almost every Mongol, and it is not infrequent to 
meet men suffering from its effect, but I do not believe that, as a whole, the 
natives are intemperate. The Mongol cheese is very palatable but usually 
it is made in such an unclean way that I discouraged its use by our men. 


The wooden pails in which the milk is drawn are never washed, and one can 
extract hair and even less desirable elements from the cheese. Nevertheless, I 
have often had to eat it in Mongolia when there was nothing else and fotmd 
that it was most sustaining. 


On the hill-slopes of the main valley and farther to the westward in the 
grasslands are many ancient remains of great archceological interest. They are 
of two kinds: one, a large circle of small stones with a low rock moimd in the 
center; the other, a rectangtdar space enclosed by upright stone slabs. The 
natives could tell us nothing about these btirial places except that they were 
"very old and had been made by people who lived long ago." Badmajapoif 
said that in the country to the west the Kozloff expedition had opened many 
similar graves which contained skeletons, together with gold and bronze 

In 1914, Douglas Carruthers, in his splendid book "Unknown Mongolia," 
discussed similar structures in the upper Yenisei basin. There he foimd 
monoliths, rock-pictures, tumuli or circular moimds, and rectangular graves. 
The two last appear to be somewhat like those we saw. He says that both 
types are burial spots, and in later years we opened several which contained 
skeletons and iron implements but no gold or bronze. Carruthers speaks of 
gigantic movmds surrotinded by upright stone slabs which he believes were 
tribal meeting places, but we found nothing of the sort; neither were mono- 
liths present in the region we visited, except in one instance. 

It is impossible to tell from the present available data who made the 
graves which we fovmd in Mongolia. Carruthers believes that those in the 
Yenisei basin probably were the work of the Uigurs. He remarks: 

"About the third century B. C. there emerged into prominence a people 
who were destined to leave their mark on the whole civilized world. Some- 
where from the far south of Mongolia, perhaps from the borders of China — 
from the present-day provinces of Shensi and Kansu — came a wandering peo- 
ple, the Uigurs. The origin of the Uigiu-s is problematic, but as far back as 
this it can be traced with fair certainty. These tribes wandered northwards, 
and eventually settled in Mongolia. Here they increased and flourished, and 
eventually spread over the Yenisei regions as far north as the Chulin River, 
until, in the eighth century A. D., their kingdom reached over the whole of 
Northern Mongolia from Lake Kossogol to the Black Irtish. ... In the ancient 
Uigtirs we have the origin of the Turkish race who later on overflowed all 
Central Asia and made an empire on the shores of the Bosphorus."^ 

' Carruthers, Douglas. 1914. "Unknown Mongolia," with three chapters on sport, by J. H. Miller. 
Hutchinson & Co., London. I, p. 51. 


We found graves, of a generally similar type to those at Tsetsenwang, 
from the frontier of China northward right across Mongolia. Those in the 
northern grasslands were much more nimierous and better constructed than 
those in the south and central Gobi. It is probable that they are of widely 
varying age and were made by people of different tribes or races, but all pre- 

Berkey and Morris discovered, between two lakes north of our camp at 
Tsetsenwang, a well-preserved and very ancient dam, half a mile long and 
fifteen feet high; also the foundations of a stone structure of several rooms. 
Some months later, near Tsagan Nor, in the Gobi, we found a similar dam and 
remains of what appeared to be irrigation ditches. These were evidently pre- 
Mongol and must have been constructed by a people who practiced agri- 

In the Yenisei basin, Carruthers mentions ancient irrigation canals and 
says that present settlers have actually opened some of them for their own use 
and found that their engineering was in no way at fault. Thus it seems evi- 
dent that western and northern Mongolia, and probably the entire country, 
was inhabited by a well-developed race, or races, long before Mongol time. 

Near one of the graves at Tsetsenwang, Shackelford picked up two pieces 
of rock, one of which contained flecks of garnet and the other gold; the latter 
would run about one hundred dollars to the ton. Doctor Berkey estimated. 
We made a careful search for the masses from which the fragments originated 
but were not able to discover them. Doctor Berkey believed that they must 
have been carried there from some distance. 


The geologists fovmd a low mountain of Jurassic sediments not far from 
camp. It contained abtmdant plant remains but no vertebrate fossils. On 
May 29, they drove fifty miles to the south and reported a further great area 
of Jurassic rock twenty miles across, and beyond it an arid desert. 


The caravan had arrived May 28, 1922, with several of the camels pretty 
well exhausted. Since we had been in a grassland country and camels cannot 
thrive on grass, they had been on very short rations. I took enough gas and 
food into the cars to let the weak camels follow without loads, and started 
the caravan for Sain Noin, one himdred and fifty miles away, with orders to 
travel slowly and feed on the way. 



We were all glad to leave Camp Tsetsenwang on May 30. It had been an 
interesting camp for the geologists from the standpoint of continental struc- 
ture, but it was rather unproductive for the rest of us. I was keen to discover 
more fossiliferous sedimentary basins, but it was evident that the general 
direction of the strata lay east and west and that we were traveling parallel 
to it. If we were to get away from the hard rocks we must turn sharply 
southward into the desert. 

The trail southwestward to Sain Noin led up and down a series of grassy 
hiUs where some of the grades were so steep that it seemed impossible for the 
motors with their heavy loads to negotiate them. Nevertheless, we went up 
without difficulty and once more congratulated oiirselves upon having selected 
the Dodge Brothers cars. Three bad stretches of dry marshland covered with 
"nigger heads" gave us an imwelcome change from the hills; then we came 
into a long valley dotted with lakelets and white with herds of sheep and goats. 
Perhaps a dozen yurts were tucked up against the hill-slope. It made a very 
pretty picture of Mongol life. 


When we were forty miles from Tsetsenwang we emerged from the hills on 
to an arid plain covered with sparse bunch grass growing on a gravelly soil of 
disintegrated granite. Off to the north lay a range of low rocky hills, cut by a 
deep canyon. We pitched the tents just at the entrance to a canyon, on the 
grassy bank of a clear stream. It was delightful to feel the soft grass under 
one's feet and to plunge into water so cold that it made one gasp. We named 
the camp "Canyon Brook." Our early stop was made at the request of the 
geologists. They had been worked well nigh to death, for the short nm of 
forty mUes had been across a complicated structixre of ancient rocks much 
folded and crushed. It was impossible to determine the relationships accu- 
rately in a few hours' study. They did not arrive in camp until very late, and 



the next day drove back over otir trail to check their work and investigate the 
surrounding hills. In our canyon they fovuid a considerable expostire of pre- 
Cambrian limestone. 

From the camp we could see, with glasses, a temple and yurts about three 
miles south of the trail . We ran toward it but found oiir way barred by an 
enormous marsh. Swans and geese were floating on the surface of a lake in the 
center, and a great flock of golden plover, numbering at least a thousand, were 
feeding among the "nigger heads." 

There were several small encampments of >airts along the edge of the 
marsh and we had a rather amusing time •with the Mongols, who had never 
before seen a motor car. The mirror, horn and electric lights were the chief 
wonders, and we created enough topics for conversation to last them for 
months by finally persuading half a dozen to ride with us for a few himdred 

On the hill-slopes and plain to the east of the marsh there are many graves 
and mounds like those at Tsetsenwang. This must have been a populous 
region in pre-Mongol days, and very probably the valley had been used exten- 
sively for agriculture. In the summer of 1925, press reports stated that Gen- 
eral P. K. Kozloflf had made important archaeological finds in this district, but 
I have no direct information concerning them. 


The small mammals at this place were interesting. Among the rocks at 
the entrance to holes I saw little piles of banana-shaped droppings which I 
could not identify. I felt certain that I was about to discover a mammal, 
new, to me at least. The next morning my traps contained a rather large 
vole-like mouse, which has since been described as a new race by Dr. Glover 
M. Allen, 1924. He has given it the name Microtus (Alticola) worthingtoni 
semicamiis and says that it "is undoubtedly a close relative of Alticola worth- 
ingtoni of the Tian Shan, nearly a thousand miles to the westward. . . . The 
fine series of this large gray microtine extends the known range of the subgenus 
northeastward into central Mongolia which may be near the limit of its dis- 
tribution in this direction." 

We found this mouse at several other camps to the westward, and covild 
always determine its presence by the piles of banana-shaped dung. It is a 
true rock mouse and the soft fur and ears are suggestive of the picas or conies, 
Ochotona. Like them it has the habit of pulling stalks of grass and flowers 
into its burrow under the rocks. The animals are partly diurnal, for I saw 
them several times in the late afternoon. They are remarkably fearless and 
curious; often they would come within ten inches of my hand. 

I trapped several conies, Ochotona pallasi, at this camp. The burrows 


were easily identified by the characteristic spherical droppings scattered about 
and the masses of vegetation drawn into the holes. On the plain at the entrance 
to the canyon I saw signs of other conies among the stiff grass, and later caught 
the animals. It was rather a surprise to find that this species lives on the plains 
as well as in the rocks. 


The next morning we followed the same faint trail on which we had trav- 
eled since leaving Tsetsenwang. It was about two feet wide and at times was 
hardly more than a shadow on the hills. Often we lost it altogether, yet this 
same trail has doubtless been in existence for centuries and runs through to 

It was amazing how much easier our traveling became if we could follow a 
trail, no matter how small. The great flat feet of a camel are natural road 
makers. They tramp down the sand and fracture the stones enough to leave 
them more easily acted upon by wind and frost and rain. Perhaps only one or 
two caravans may follow a trail during a season, or even several years may go 
by without the passing of a single camel, but still the trail exists. There were 
many occasions when it would have been impossible for us to cross a sandy 
area with our cars had we not discovered a caravan track. 

A network of trails crosses Mongolia. The most important run from 
Kalgan, Kweihwating and Paotowchen on the Chinese frontier, north to Urga, 
and northwestward to Uliassutai and Kobdo. These have been in existence 
for hundreds or thousands of years. They are the trade routes which connect 
India, Russia and China; the routes which the great silk caravans followed, 
even before the days of Kublai Khan and Marco Polo. They are being fol- 
lowed to-day in the same way, but by camels carrying different products. 

The great routes are marked by wells, sometimes ten miles apart, but 
often as much as fifty or sixty miles, depending upon the coiuitry and the 
depth of the water-level. From the "main lines," branch trails nan in all 
directions. They may go a few miles to a temple or may end seemingly in 
"mid-air." Usually, when they have no apparent excuse for being, it means 
that at some time during the year the spot where they end is a favorite Mongol 
camping ground. When the use of motors becomes general, as must inevitably 
happen, and all the passable trails have been accurately plotted, the map of 
Mongolia will look like a spider's web. 

The average Mongol knows comparatively little about the trails except 
those in his immediate vicinity. Each year he moves perhaps half a dozen 
times, but always in a certain region prescribed by his local prince or chief. 
He cannot go beyond the area that has been allotted him or he will infringe 
upon the grazing rights of others. He knows the trails within that region like 


a book, but unless he has happened to make a joiimey to some far-distant place 
he can tell you little about other sections of the country. 

The people from whom to obtain real information are the Chinese traders. 
Year after year they cross Mongolia, taking tea, cloth and tobacco to far-dis- 
tant points in Central Asia and returning with wool, hides, furs and ponies. 
At present Mongols do but little caravan trading on their own account, but 
the Chinese merchants will often hire a few to assist with the camels. 

We could always recognize a Chinese caravan at a distance because the 
Chinese invariably use camel bells, whereas the Mongols almost never do. 
Again, the Chinese caravans usually travel with the camels in single file in 
groups of twenty, each animal tied by its nose-rope to the load of the one in 
front of him; the Mongols often lead the caravan in a compact mass. The 
Chinese herders in charge of each group of camels usually walk; the Mongols 
always ride. 

The picturesque caravan traffic in Central Asia will persist for many years 
even after the inevitable railroad has crossed the desert. There are many 
isolated regions, not rich or important enough for railroads ever to reach, 
from which the products are too bulky to be carried by motor car. Therefore, 
I believe that this transport of the Middle Ages will continue to exist to some 
extent even when the twentieth century has drawn to a close. 


The first twenty-five miles westward from "Canyon Brook Camp" was 
what Morris called a "geological nightmare." We fervently echoed his senti- 
ments from another standpoint, for it certainly was a "motorist's nightmare" 
as well. The tiny trail wound in and out among rocks and over ridges on some 
of the worst terrain we encountered in all Mongolia. The hills were covered 
with a chaos of jagged stones. In every direction masses of granite, like the 
vertebral colimms of huge dinosaurs, cut the surface, and enormous boulders 
were heaped one upon the other as though they had been a giant's playthings. 
It was a dreary country. Only a faint shading of yellow-green vegetation 
relieved the monotony of dull brown. There were no yurts and we met only 
three Mongols. They were huddled under a wretched little tent beside the 
trail and gave the crowning touch to the picture of desolation. They were 
bound, they said, for a great lamasery far to the southward in the Altai 


After leaving thfe rocks the trail took us into a dry marsh among "nigger 
heads," which only changed the character of the bad going. Just beyond the 
marsh our trials were ended. A series of rounded hills, and then a gravel 


plain almost as smooth as a tennis court, descended very gradually to the 
Ongin Gol. 

I had drawn a mental picture of a gently flowing river with grassy banks, 
shaded by willow trees. The reality was a disappointment. A shallow stream, 
split into three small branches which wander rather aimlessly through the 
gravel plain, is the Ongin Gol at this point. It flows southeastward from the 
Khangai mountain range, which forms the Arctic Divide farther to the north. 
Where we crossed it, the river occupies a valley several miles wide, within 
which is cut a still lower trench about a third of a mile in width. The stream 
occupies only a small part of the inner valley, though it is evident that the river 
has flowed in other channels in past times. Not a tree could be seen, but there 
was grass, and we camped on the west bank early in the afternoon. 

The geologists had lagged far behind, and when they arrived their Chinese 
chauffeur drove recklessly into the widest branch, stuck fast between two rocks, 
jammed on full power, and promptly snapped the rear axle. It required hours 
of hard work with the block and tackle to get the car out and to replace the 
broken part. 


I was surprised that there were no yurts beside the river where the fresh 
grass offered such excellent feed. Our Mongols could not explain it, but when 
I sent Tserin over to a nearby ridge to shoot some sand-grouse for dinner, he 
came back with two wretched Mongols who related a sad tale. 

Only a few months earlier there had been an encampment of seventeen 
yurts almost where our tents were pitched. It was a peaceful Mongol village 
numbering about fifty individuals. One day a Chinese caravan from Ulias- 
sutai, on the way to Kweihwating, laden with sable skins and other valuable 
furs, had camped beside them. During the night a party of Russians arrived 
and slaughtered every living soul, men, women and children. It was "spurlos 
versu7ike?i" ; so far as they knew, not a hiiman being was left alive to tell the 
tale. Then the Russians drove off the camels with their loads of furs, and all 
the Mongols' sheep and goats. Even the dogs were shot before they left. 

The two Mongols who told us the story were away the night of the mas- 
sacre and had arrived just in time to witness the destruction of their homes 
and families. From the shelter of the ridge they had watched it all, and since 
that time had lived half-crazed among the rocks, unable to tear themselves 
away from the bloody spot. Unfortunately, there was every reason to believe 
that the tale was true; it was only one of many similar events that marked 
the frightful winter of 1921. The things I had heard in Urga, of the "Bloody 
Baron" and his awful vengeance upon the Jews and Bolsheviks, recalled the 
worst days in the history of the inhiunan "Ivan the Terrible." 



We awoke, on the morning of June 2, to a lowering sky and certain indica- 
tions of rain. I knew that it was imperative to get the Expedition away at 
once, for two miles of not too dry swamp land lay between us and high ground ; 
a few hours of rain would turn this into an impassable morass. 

The car was repaired, and camp was broken at eleven o'clock. We nego- 
tiated the "nigger heads" successfully and, as usual in Mongoha after particu- 
larly bad terrain, we then fovmd ten mUes of level plain. Half a dozen herds 
of gazelles were feeding near the trail and I killed two for meat. The firing 
not only secured the animals but frightened nearly to death the inhabitants 
of two yurts, from whom we wished to obtain information concerning the 
trail. The men leaped upon their ponies and dashed for a rocky outcrop, 
leaving the women to their fate. As Shackelford pertinently remarked, the 
Mongol motto is "Save the men, to hell with the women and children." It 
took considerable quiet conversation by our Mongols to assure the poor women 
that we were friends and would not harm them. The slaughter at the Ongin 
Gol was fresh in their minds, and the sudden appearance of strange machines, 
the like of which they had never seen, combined with rifle shots, made them 
certain that a horrible death awaited them. Eventually they told us that the 
tiny trail we were following went to Uliassutai, and that we must cut 
across country northward to a big road which reached the Sain Noin mon- 

We had hardly gained the trail before a storm broke with terrific violence. 
We covdd only back the motors into the wind and keep as dry as possible. A 
deluge of rain and hail poured from a night-black sky Ut by continuous light- 
ning. The crash of thunder made it impossible to talk. The storm raged for 
half an hour, and when we saw the brown rivulets cut their way down the 
sparsely covered hillsides we began to appreciate the erosion effects of such 
torrential rains. 

Forttmately we were only a short distance from Sain Noin, and pitched 
camp in the valley of the Tarmil Gol, a tributary stream which flows north- 
ward to the Ongin Gol. We called it "Rainy Gulch Camp." The valley was 
well populated, for Mongolia. Opposite camp a village of a dozen yurts lay 
on a hill-slope. We made friends with oiu- neighbors at once, and in return for 
baskets of dry argul gave them cigarettes and empty gasoline tins. Bottles 
are such a desideratum that we kept them for very special favors. 

Our camp at Rainy Gulch was at a higher elevation than any we had 
reached up to that time. The aneroid readings indicated more than six thou- 
sand feet. The moimtains which we could see from camp form the divide 
between the drainage flowing toward the Arctic and that flowing southward 
into the desert. 






The following day we drove to the monastery, five miles away. Because 
of its beautiful setting it is the most impressive of any that I have seen in 
Mongolia. The summit of a long grassy hill is cro'wned with an enormous and 
well -constructed obo. It consists of a circular base of stones, with a secondary 
tier and a conical center decorated with prayer flags, bits of cloth and branches. 
As this is a very sacred spot to the Mongols, Badmajapoff, who was with us, 
warned us not to approach too closely. Below the obo on the green plain, the 
golden spires and upturned gables of the temples glistened in rainbow colors. 
Just beyond the monastery the Ongin River has cut its way through a table of 
solid rock, and in the distance wave upon wave of snow-crowned mountains 

The temples are in the center of the "city," with the tiny wooden houses 
of the lamas spread out on each side like great wings. A palisade of impeded 
logs siorrounds the entire monastery, and most interior compounds have their 
own enclosures. The temples are interesting and show three distinct types of 
architectiu-e. The first has the squat rectangidar Tibetan base, but the upper 
half is a typically Chinese, pagoda-like structure. Behind it stands a temple 
of pure Tibetan construction, square, massive and flat-topped. Enormous 
troughs of weathered wood, like rain spouts, extend out at intervals from near 
the roof. The walls are white, bordered by black with red bands; the comer 
decorations are of gold leaf. A third temple is of Chinese architectiu"e through- 
out, roofed with green glazed tiles. There are ten temples in the monastery. 
The usual whitewashed shrines and prayer wheels are particularly numer- 
ous; the central shrine has the lower part plated with sheets of tin from 
Standard Oil cans, and the upper half covered with pxxre gold leaf. 

The Khan's winter palace is in the northeast comer of the lama city, and 
with his private temple is siurounded by high palisades. The Khan was then 
only ten years old, and his uncle, a high lama, was acting as regent. Later we 
visited the regent at the Hot Springs. 

We called upon various officials at the monastery, after Badmajapoff had 
visited them, to explain our objects, and were most cordially received. There 
was much more order and discipline among the lamas in this monastery than 
in any other which we visited in Mongolia; moreover, the lamas themselves 
and the surroundings are exceptionally clean. 


Although we had no hope of finding fossiliferous sedimentary basins in 
this region, we were obliged to wait until the camels arrived; nevertheless, 
there was much of interest for all the members of the Expedition. The geolo- 
gists, especially, were busy from sunrise until late at night, investigating the 


complex rock structure of the valley. They found that for a long distance the 
graywacke series, which is much folded, jointed and crushed, forms the hills 
and mountains to the northeast of the monastery as well as the more distant 
higher mountains. Indeed, it is the dominant formation of the Khangai Range. 
In the vicinity of the lamasery there are several miles of basaltic flows, which 
have been poured out into the bottom of an older valley by late volcanic action. 
Since that time erosion has continued and the stream is again channeUng its way 
through the volcanic material. 


On the northern slopes of the mountains, just beyond the lamasery, were 
patches of larch trees, and I decided to move camp sixteen miles to the north 
so that we might have an opportunity to study the forest conditions. A day 
of prospecting to locate a suitable spot took us across the Ongin River and up 
a long valley north of the monastery. As we were about to cross the stream, a 
great herd of yaks, driven by Mongols, came down the opposite bank. They 
were the first yaks any of the men, except myself, had ever seen. Each one was 
trailing a log on either side, and the great shaggy brutes, with their enormous 
tails and the long belly-hair reaching to the ground, made a splendid picture 
as they dashed into the water. They were very wild, snorting and grunting 
whenever they came near us, and the IMongols had a lively time to handle 

Most of the animals were piire-blooded yaks, but a few showed by their 
lighter build, shorter hair and smaller tails that they were "yak-cows" — a 
cross breed between yaks and the Mongol cattle. This cross appears to make 
very good work animals and I have seen many drawing carts in Urga. They 
are not so wild and difficult to handle as are the pure-blooded yaks. Yaks are 
used extensively in the forest region of northern Mongolia, and even as far 
south as the eastern Altai Mountains. We found a few at Ikhe Bogdo, but 
none was seen in the desert region, for yaks are high altitude animals. 


Ten miles north of the monastery our trail led abruptly to the crest of a 
sharp divide, eight thousand feet above sea-level. A magnificent view lay 
before us. In the bottoms of deep, heavily forested valleys, shining threads of 
water wound snake-like to the river far away to the east. Fifteen miles away 
was the Arctic Divide, formed by range after range of snow-crowned mountains. 

On the slopes of a side valley, in a beautiful patch of larch forest, we found 
an ideal camp site. A considerable yvu-t village lay on the opposite grassy hill- 
slope, and a spring of sweet, ice-cold water ran off into a tiny brook thickly 
margined with alders. We returned to the camp at Rainy Gulch thrilled with 


anticipation of the days to come. Badmajapoff had decided to visit the Hot 
Springs at Arishan, fifteen miles away, while we were at the Forest Camp. 
He was sioffering from rheumatism as a result of tortures inflicted by Chinese 
soldiers during the terrible winter of 1 920-1 921. Colgate and three of the 
other men drove over with him, and found that the Prince Regent was also 
taking "the cure" at the Holy Mountain. They left Badmajapoff comfortably 
quartered in a yurt, where he was to remain imtrl we picked him up on our way 
southward after the caravan had arrived. 


The night before we broke camp all the Mongols of the little village across 
the valley gathered at our tents to see the preliminary packing. They were 
genuinely regretfiil at our departure, for we had brought much of interest into 
their tmeventful lives. In particular, one little girl twelve years old watched 
every detail with the greatest attention. When it was almost dark her father 
said, "It is time to go home now." 

"No," said she, "I don't want to go. I want to look at all these things as 
long as I can. I'll never see anything like them again." 

The Mongol children are exceptionally bright and are advanced in knowl- 
edge far beyond their years. This is due to the self-reliant life which is forced 
upon them almost from the time that they have learned to walk. Certainly 
they are examples of the "survival of the fittest." They grow up as best they 
can, with no attention to the most ordinary rules of health, cleanliness or 
diet. From the moment they stop nursing they eat and drink whatever comes 
in their way. If they are iU, the only medical services are the prayers of the 
lamas. They must become hardened to the greatest extremes of temperature. 
Often I have seen babies, only two or three years old, running about stark 
naked, outside the yurt in a bitter wind, when I was shivering in a fiir coat. 

When a child is about four years old it is taught to ride a pony. No 
sympathy is wasted if it falls off; it is only put into the saddle again, and some- 
times tied in place. At five or six the children begin to do their bit at herding 
sheep and goats; a few years later they graduate to the care of camels and 
ponies. This necessitates long hours in the saddle and often nights alone on 
the desert. They must learn self-reliance long before their time. They have 
accumtilated most of what their elders have to teach them by the time they 
are sixteen, and from that age onward they make little mental progress. 


On Jtme 7, we left Rainy Gulch for the camp in the forest. It was by no 
means easy for our heavily loaded cars to negotiate the trail, because of the 
steep grade, soft ground and mud holes. With careful handling, however, all 


the motors arrived without accident and we camped at noon just within the 
edge of a beautiful forest. 

There was not a breath of wind, the sun lay warm and bright, and the 
air was sweet with perfiune from the larch trees. Beneath our feet was spread 
a gorgeous carpet of flowers: yellow buttercups, forget-me-nots, pxirple iris, 
phlox and a dozen others of rainbow colors. The floor of the forest was thickly 
carpeted with dead brown "needles," and the larches already showed new buds 
of brilliant green. Thus, suddenly, we had come into entirely different physical 
conditions. We were camped on the borderline between the forest and grass- 
lands, and almost at the divide separating the drainage of the Gobi from that of 

The forest edge is a zone of conflict. There were only two stages in tree- 
growth represented in the margin of our grove — seedlings and saplings not more 
than twenty-five years old, and ancient trees, which, judging by their annual 
rings, were about three himdred and sixty years of age. This would indicate 
that there was a period of favorable climate when the seedlings were able to get 
a foothold, followed by a long period of comparative drought on this side of 
the divide, and a more recent period of moisture. The woodland area in which 
we were camped was the most southern of all the forest, and the trees have just 
been able to maintain themselves in their battle against climatic changes. The 
larch is apparently the only species of tree that has been able to exist. On the 
other side of the divide, birches are numerous and there is some spruce and pine. 

Although there is a carpet of moss under the trees, it is much less thick 
than that on the northern slopes of the divide; there, the moss is like a mat- 
tress, and imdemeath it the rich black earth is always damp or wet. Because 
of this lack of moisture there are comparatively few small mammals on the 
southern slope of the mountains; in fact, even such usually abundant species 
as Microtus are scarce and we caught no insectivores. Across the divide the 
earth was honeycombed with small mammal runways, and our traps yielded 
lemmings, Myopiis, red-backed mice, Evotomys, voles, Microtus, and a fauna 
typical of the northern forests. 

WhUe the trees are in patches, to the south of the minor divide on which 
we were camped, the view from the ridge to the north showed an almost 
unbroken line of forest. Streams, bogs and marshes are present in every valley, 
and water trickles down the hillsides xuider the mossy covering. To those who 
know MongoUa only as a region of grasslands and desert, tales of the beauti- 
ful northern forests soimd well-nigh unbelievable. I knew its forests well, for 
in the summer of 191 9 I traveled through the great forests north of Urga. It 
is a land of splendid trees, of luxixriant meadows and gorgeous flowers, as differ- 
ent from the desolate Gobi only a few miles to the south as is day from night. 

The fatma is that of the Mongolian-Siberian Life Zone, which is roughly 


delineated by the southern edge of the forest. Moose, wapiti, roe deer, wild 
boar, musk deer, foxes, lynx, sables and sqvdrrels are to be foiuid in greater or 
lesser abundance. Such typical Siberian Zone birds as the capercaillie, Tetrao 
urogallus, black grouse, hazel hen and ptarmigan foUow the trees almost to 
the border line where they meet the grassland fauna including marmots, 
spermophiles, jerboas, field voles, antelope, bustards and many species of 
plains birds. 

Owl camp was seventy-seven hundred feet above sea-level and even 
though it was mid-June the nights were cold. Usually the temperature dropped 
to the freezing point, but the days were delightfully warm and I cannot imagine 
more perfect weather. We all enjoyed this sojourn in the forests, but, except 
from the standpoint of zoology and geology, it was not very productive. The 
geologists found that the entire Khangai mountain region was a continuation 
of the great gray\\'acke series, with nothing to break the monotony except an 
infaulted Jurassic remnant and a basalt valley-fill near the Ongin River. 
From the summit of the minor divide near camp we could see several splendid 
cirques, representing glaciation of an alpine type, in the next range. Berkey, 
Morris and Colgate made an attempt to reach them, but the trail proved to be 
impassable for the car and they had to spend the night in the open after an 
exhausting day. Travel in the forest region of Mongolia is impossible for any 
wheeled vehicle except an oxcart, due to the mud which lies in the bottom of 
every valley. 


On June 13, Merin, our caravan leader, rode into camp. He reported 
that the camels were awaiting us with Badmajapoff at the Hot Springs. All 
was well with them except that some of the animals were weak from lack of 
food ; when crossing an arid stretch they had had nothing to eat for three days. 
The caravan had been attacked by small parties of brigands five or six times, 
but the robbers had been driven off by rifle fire from several of our men and a 
Mongol soldier who was attached to the caravan. 

We left the forest camp with regret, on June 15, and faced southward 
toward the desert. Not until we rettimed to China should we see trees again 
or have more evenings about the great log fires; henceforth the only fuel 
would be bits of argul. Nevertheless, we were anxious to be off, for since the 
first thrilling days of Iren Dabasu no fossil-bearing sediments had been dis- 
covered. We had, however, learned much about the peculiarities of the coun- 
try and the continental structure; we knew that the northern grassland region 
through which we had been traveling was one of hard rocks, generally nmning 
east and west, and that if we were to find eroded sedimentary basins we must 
turn sharply southward. 


Just as we were crossing the Ongin River, near the temple, an enormous 
drove of sheep, cattle, goats and ponies appeared, and behind them fifty camels, 
some loaded with ytuts and household goods. It was a small Mongol village 
moving from the grasslands to the summer camps in the mountain valleys. 
The annual allotment of grazing regions had taken place and this was "moving 
week" for the Mongols. 

Our friends at Rainy Gulch had transferred their village to a valley nearer 
the temple, and as our cars passed they gathered by the trail to bid us fare- 
well. Nearing the sacred mountain of Arishan, we saw otu" caravan camped 
on the summit of a rise directly opposite the Hot Springs. 

Badmajapoflf delighted us with the information that he had heard from 
the Mongols of a region, eighty miles to the south, where fossil bones were to 
be found; bones "as big as a man's body," they said. We thought it was 
merely native exaggeration, of course, but discovered later that they were 
not so far wrong, after all. 


The Hot Springs is an interesting place. The water gushes out from under 
a rocky ledge, spreading into a multitude of streamlets as it runs down the 
slope. Just above the spot where the spring emerges from the hillside, an obo 
has been built, in the shape of a semicircular rampart. Inside the rampart 
there is a tiny altar, bearing three flat stones upon which are etched pictures 
of Buddha. Scores of silken scarfs, many faded into a imiform dvdl white 
from the rain of countless storms, drape the altar. Some have been whipped 
to ribbons by the wind ; others are new and as blue as the sky above the shrine. 
These are the offerings of pilgrims who have come to bathe and to drink the 
sacred water. 

The shrine is the "holiest of holies" and we were asked not to step within 
the obo circle. A lama showed us a nest of vipers beneath a stone; indeed, 
the whole vicinity of Arishan swarms with vipers. When they crawl into the 
tents and yurts, the natives gently shoo them out the doors. No life may 
be taken at this place; not even an insect or a worm is killed if it can be 

At intervals along the various streamlets, pools have been built in the 
rocks; over each a tent or yurt is placed. The pool farthest up the hillside 
near the shrine is the place of honor and is reserv^ed for the Prince, or Ta Lama 
of Sain Noin. Badmajapoflf had a bathing tent next that of the Prince and, 
after a brief inspection of the place. Granger and I went in for a bath. The 
water was crystal clear and only slightly sulphurous. 

A cold spring flows out near the hot one and this has been diverted in tiny 
branches to each bathing pool; thus one can control the temperature of one's 


bath by damming up or widening the cold stream. The actual temperature of 
the water where it issues from the ground is 127° F. 

The motmtain, Arishan, from the base of which the hot water gushes 
forth, is aU sacred ground. It is little to be wondered at, this worship of the 
spring. Imagine a caravan in the bitter days of winter, when the wind cuts 
like a white-hot brand, winding over the hills and pitching its tents on the plain. 
When the loads have been lifted from the tired camels, the half-frozen nomads 
walk across the valley to find a spring of steaming water bubbling from the 
frozen earth, offering health and comfort. What better evidence cotdd there be 
of direct connection with supernatural powers! 

Doubtless this motmtain has been the object of pilgrimages through 
ages. Indeed, we foimd evidences of an ancient edifice in a chaotic mass of 
rocks below and at the side of the spring. They appeared at first to have 
been washed there by some violent flood, perhaps from a cloudburst on the 
mountain slopes, but upon closer examination we discerned lonmistakable 
evidences of an orderly arrangement of the large blocks ; moreover, some were 
foreign to the mountain at the foot of which they rested. There is little doubt 
that they formed the base of a huge shrine or ceremonial biiilding at some time 
in the far, dim past. Perhaps those pre-Mongol people whose graves we foiond 
in the grasslands had constructed an altar here! Perhaps it was before their 
time and they themselves had built upon the ruins left by still older tribesmen. 
Thus, the present shrine above the spring has been built by the Mongols upon 
a foundation of these ancient blocks. 

Regarding the scientific reason for the Hot Springs, Professors Berkey and 
Morris, 1927, have said: 

"The holy spot lost none of its charm for us because it happened to fit 
oiir purpose to study the mystery itself, for it is one of the type geological 
localities of northern Mongolia. 

"We found a great fracture in the earth's crust at the base of the motm- 
tain, and the strata have been displaced along it so that one side has dropped 
down several thousand feet. Taking advantage of this weakness, volcanic 
lavas have attempted to break through the fault zone, but their energy died 
out before much of an outbreak was accomplished. They came so near the 
surface, however, that even to-day one can find a few fragments of volcanic 
rock lying about on the ground, and probably there is much more below. 
Now, the rain that falls on the adjacent hills and sinks into the ground finds its 
way into the fractured zone, following along the hillside and under the valley 
till it encounters the lava below, which must still be hot. Heated by the 
lava, the waters rise of their own accord, issuing as hot springs."^ 

'Berkey, Charles P., and Moms, Frederick K. 1927. "Geology of Mongolia," Natural History oj 
Central Asia, Amer. Mus. Nat. Hist., II, v-xxxi, pp. i-47,=i. 


The geologists had an extremely busy time at the Hot Springs, because 
they decided to make a detailed map of the locality. From daylight until 
long after dark they worked so feverishly that I was seriously concerned for 
their health. 


On our first day at Arishan we called upon the Prince, or Ta Lama. He 
made an exceedingly pleasant impression upon all of us. About thirty years 
old, rather small in stature, but well-formed, he has a delicate, sensitive face 
and fine eyes. His skin was very white — much lighter in color than our own 
faces, which were tanned by the sun and wind. Doubtless there are some 
lamas who take their religion seriously, from the purely religious point of view, 
but I have met very few of them. The Prince was a shining exception. Not 
only does he conform to the Buddhist principles in being a true celibate and 
avoiding wine and tobacco, but an inner goodness "shines" from his face. His 
expression is rather sad but singularly gentle, and there is a quiet dignity in 
his carriage and in every action. When in repose he imconsciously assumes a 
Buddha-like attitude and one feels that he has attained absolute mental com- 
posure. He smiled sometimes but I never saw him laugh. We asked him to 
dinner at our camp, and I noticed that he watched the ways in which we 
handled our knives and forks in order that he might not make inistakes. 
The ordinary Mongol, of course, eats with chop-sticks or his fingers. 

The Prince was interested in everything — not a childish curiosity, but a 
real desire to understand the extraordinary things in our equipment. A car- 
bide lamp delighted him and he comprehended at once how the gas was formed. 
He never tired of investigating the internal anatomy of the motors, and the 
only time that I saw his habitual calm disturbed was when he was allowed to 
steer the car which carried him back to his yurt. For a few moments his face 
wore an expression of intense delight. I have written somewhat at length 
concerning the Prince because he is by far the most remarkable Mongol whom 
I have ever met. 


The valley of the Sacred Mountain swarmed with antelope and at this 
date (Jtme 17) they had just assumed their fuU summer coats. They were 
all of the species Procapra gutturosa, because the desert gazelles never range so 
far into the grasslands. The valley was a very populous place during our stay 
there. As I have remarked, this was "moving week" for the Mongols, and 
dozens of families, with thousands of sheep, goats, cattle, ponies and camels, 
passed northward to the mountains. Usually they arrived early in the after- 
noon and were gone at daylight in the morning. 










A large family on the march is an interesting sight. The great herds of 
goats, sheep, cattle and ponies are driven along rapidly in front by moiuited 
herders. The camels travel in the center bearing the yurts and household 
goods packed in neat bundles; behind ride the men and women on camels and 
ponies. The babies are packed in wicker baskets on either side of a camel, 
and it is most amusing to see their little heads and bright eyes peering down 
from the huge animal. Since the transient visitors were on the move shortly 
after simrise, it was difficiilt for Shackelford to record all he wanted in motion 
pictures. They must always arrive at their new camping spot early enough 
to give their stock several hoiirs of grazing before being gathered in for the 



On June 15, we sent the camels southward to the edge of the desert, where 
the feed was better suited to these strange animals. The spot was on the route 
which we wovdd follow to the fossil locaUty reported by the Mongols. We 
ourselves left the Hot Springs three days later. We drove down the valley 
imtil we reached the faint path which marked the Uliassutai trail. All Mon- 
golia seemed in motion, for every mile we passed newly pitched yurts, with the 
sheep and goats gathered in closely packed units or scattered over the plains. 
Antelopes, too, had joined in the migration toward the mountains and were 
moving northward in great herds. 

The trail led us up and down a series of grassy hills, each being steeper than 
the last, until, after an almost perpendicular climb, we started down a long 
gentle slope where the cars coasted for miles. It took us to the bottom of a 
wide valley in which a single yurt was pitched beside a tiny stream. For 
twenty miles the trail had been deserted, and we realized, from the rapidly thin- 
ning grass, that we were nearing the desert's edge. Crossing the stream, we 
turned straight south down the valley, and at five o'clock in the afternoon, 
from the summit of a hill, we sighted the blue tent and grazing camels of our 


The trail had taken us over a coimtry composed mostly of graywacke and 
granite rocks. Owr Mongols had camped at the base of an extraordinarily 
picturesque moimtain range of bathylithic granite. The bare, jagged peaks 
showed dim and ghostly against the sky, in the simset haze; the terrain was 
fine gravel, studded with clumps of long grass, stiff and hard. Already camel 
sage and low thorny bushes had begun to appear. It was in truth the edge of 
the desert, even though we were only fifty -four miles south of the Hot Springs 
in the midst of the grasslands. 

Merin said that the feed was excellent for the camels — sagebrush and 



thorns ! The camel is in truth a relic of the Pleistocene, and his tastes are just 
as peculiar as his appearance. In the midst of sweet green grass he languishes 
and grows thin, but put him among the thorny desert vegetation and he is 
happy. His htimps and belly grow fat and round, and he stores up nourish- 
ment enough to last him for months of scanty food. All our camels were in 
good condition, but badly in need of rest and an opportunity for uninterrupted 
grazing. The felt strips, which are woimd aroimd and aroiuid the himips to 
form the "saddle," had not been removed since March 20, except for the 
examination of some individuals with sore backs. Therefore, I told Merin 
to remove them all, and that we would let the camels feed while we prospected 
the svuTOunding coimtry in our cars. 

Our camp in the basin, which the Mongols called "Gorida," was only a 
short distance from the Arguin Gol, a shallow stream. A mile away was a 
small saline lake, stirrounded by enormous "nigger heads," and three miles 
east we covild see a much larger body of water shining in the sunlight. Several 
good springs near the river afforded drinking water. 

Both lakes were abundantly inhabited by breeding waterfowl, bar- 
headed geese outnvimbering all the rest. The young were about one-third 
grown, and I spent a most interesting hour watching the parent birds piloting 
their broods to hiding places on shore at signs of danger. The little fellows 
were adept at concealing themselves in the thorny vegetation which covered 
the gigantic "nigger heads." The female geese would not desert their young 
under any circumstances, but the mallard ducks were by no means as 

As far as I can remember Gorida was the only camp in Mongolia where 
we were really annoyed by sand-flies. These insects quickly drove us into the 
tents, where they would not follow if a canvas floor-cloth had been spread. 
We could sit in perfect comfort a few feet inside the tent door, whereas out- 
side it was most vmpleasant. The absence of biting insects is one of the most 
delightftil things about camp life in the Gobi; such pests are present in so 
few places that they can virtually be eliminated from one's calculations. 
In the northern forests, however, this does not hold true; there flies and mos- 
quitoes are ordinarily troublesome. 

I was most impressed by the abundance of bird and animal life here at 
the edge of the desert, in comparison with the forest which we had left so 
recently. There the woods were nearly deserted except for a few species of 
birds, which were not abundant. Here, birds were everywhere. Our traps 
held so many small mammals that the three taxidermists were busy every 
moment; at the forest camp I set fifty-one traps one night and not one was 
sprung the next morning. Of course, this was partly due to the fact that the 
woodland on the south side of the divide, where we were camped, was much 


drier than that on the northern slope. Nevertheless, I have found that, as a 
rule, the evergreen forests in northeastern Asia do not have abvindant bird or 
animal life, as compared with the deserts. 


Our tents were pitched a hundred yards from the Kalgan-Sair Usu- 
Uliassutai trail. This is an ancient highway, which, after the Chinese had con- 
quered the Mongols, had become one of the most important routes across the 
desert. It was known as the "Post Road." Branch trails leave it for Urga and 
Kobdo,- and subsequently we found that for much of its way wells have been 
dug every ten miles. It was along this trail that Chinese officials traveled on 
their visits to the important administrative centers of Mongolia, and at sev- 
eral points the ruins of Chinese mud rest-houses could be discerned. 

This Sair Usu trail is still a trade artery through the Gobi. From our 
camp we could see the blue tents of a great Chinese caravan, a mile away. 
They remained there for two days, resting and allowing the camels to graze. 
The night before we left for the south the caravan took to the road. Just as 
twilight was fading into night we heard the mellow dong, dong, dong of deep- 
toned bells, and saw a long line of curving necks and huge bodies loom against 
the eastern sky. They came slowly toward us, two himdred camels in groups of 
twenty, a phantom-like procession. These grotesque bodies, moving almost 
without a sound as their padded feet trod the sand, passed and disappeared 
toward the glow of the western horizon. For me there is no more weird and 
ghostly sight than the silent progress of such a caravan through the desert 

I talked with two of the Chinese. They were Mohammedan merchants 
from Kweihwating, in Shansi, bound for Uliassutai with tea, cotton cloth and 
tobacco. Five months later they would return with skins and wool. Perhaps 
they might also drive home a herd of ponies to sell to the farmers in north 
China. Year by year merchants make these journeys to the heart of central 
Asia, exactly as their ancestors did a thousand years ago. The Chinese eagerly 
asked us for news, and we were able to give them some, for they had left 
China weeks before we had started from Kalgan. 


From the granite crags near Camp Gorida, we could look southward across 
a great basin to two prominent mountain ranges. The nearer, almost thirty 
miles away, was Moimt Uskuk; far beyond and towering above it was Baga 
Bogdo, of the eastern Altai. 

The great Altai system extends southeastward into the Gobi. Toward 
its eastern end it becomes lower and less rugged and breaks up into partly 




d I 


The surface of this lake h.-i(l a crust of pure salt an inch thick, in 1922. 



isolated ranges and spurs, which gradually lose their identity and merge into 
the rolling desert. Baga Bogdo, the "Lesser Buddha," is one of these partially 
separated groups. The snow-crowned summit of the highest peak lay against 
the background of blue sky, glittering like a piu-e white diamond. In the great 
desert basin, and at its feet, we hoped to find fossil-bearing sediments. 

We had had delightful weather, but on June 20 a storm blew up and the 
temperatvu-e dropped to freezing, even though it was June. All of us pulled 
fur coats out of our duflfle bags, and shivered in the gusts of hail and snow 
which whirled into the tents. The weather changes of Mongolian spring are 
kaleidoscopic. A day of hot weather, when one is comfortable only in the 
thinnest of clothes, may be followed by a drop of thirty or forty degrees of 
temperature in a few hotirs. One cannot be really sure that summer has come 
to stay until the first of July. 


The Mongols at the Hot Springs of Arishan had told us that at Gorida 
we would be able to find a guide who could take us to the locaUty where fossils 
had been reported. Colgate and Badmajapoff spent all day of June 19 driving 
about the country, trying to find a certain "rich man"; he could not be dis- 
covered, but at last they came to a village of six ytu-ts and were directed to the 
poorest man in the whole region. His yiut consisted of a few pieces of felt 
tied together; his worldly possessions included one wife, one horse, one sheep 
and one goat. Still he appeared well-fed and happy, although, like "Gunga 
Din," his clothes consisted of nothing much before and rather less than half of 
that behind. He appeared to know a good deal about the country into which 
we wished to go, and readily agreed to come with us if he cotdd borrow three of 
our camels to take his wife and possessions to a friend ten miles away, who 
would look after them in his absence. 

Our guide directed us straight across coixntry, where there was no trail, 
toward the base of Mount Uskuk. It was his first experience in a motor car — 
indeed, he had never even heard of such a thing before — ^but he was prepared 
to enjoy himself to the fullest. Granger gave him a cigar, which was also new 
to him, and he settled back in his seat beside me with an expression of the 
most sublime delight upon his face. 


About fifteen miles from camp we crossed a low ridge, into a considerable 
basin, and saw a brilliant white lake glistening in the sunlight. It proved to 
be a salt lake — or rather, a lake of salt, for the surface was a sohd crust of 
salt, several inches thick. Near the eastern end, half a dozen Mongols, with as 
many camels, were gathering salt. We ran along the shore on sand as hard 


and smooth as a sea beach and halted for tiffin near the Mongols, for the place 
was too picturesque to pass quickly. 

Shackelford set up his motion -picture camera and fovmd an excellent actor 
in a little bowlegged Mongol. We made him understand that we only wanted 
him to do his usual work, without looking at the camera. With a most serious 
air the old fellow grasped his pail, walked resolutely through the mud to the 
fresh crust away from the shore, broke off enough salt to fill his bucket, and 
came in again to spread it on the ground to dry. 

Our movie star had with him an attractive Mongol girl , who was left in 
charge of the camels. When the photographing was ended we presented the 
old man with an empty gasoline tin and gave the girl a dozen pictures of 
actresses, torn from an American magazine. She was too embarrassed to do 
more than smile faintly, but the way she clasped them to her breast while 
holding the camels' rope with the other hand showed how much she valued 
them. A few moments later, when our motors frightened the camels and they 
began plunging in all directions, she dropped the papers and a gust of wind 
swept them over a hill. For a moment she stood irresolute, looking at the 
camels legging it down the beach, then dashed after the precious pictures, 
leaving the animals to go their way. 

The salt was as pure and white as though it had been refined for table use. 
The thick crust lay on a bed of soft black mud. I never have seen another 
deposit of salt so beautiful an>'-vvhere in Mongolia. I was surprised to find how 
comparatively little salt the Mongols use. In most parts of the desert it is not 
difficult to obtain, for usually the small interior lakes are more or less saline, 
but the natives prefer their meat, butter and other animal products unsalted. 
They like sugar very much, but I never foxmd a Mongol who had any in his 
yurt, and it can be obtained only from passing Chinese caravans; sweet choco- 
late is not much appreciated even by the children. 

In regard to our salt lake, Berkey and Morris, 1927, have remarked: 
"The little dry lake has no outlet, but there must have been a time, not long 
ago, when this was simply a portion of a valley through which a stream ran, 
and there must have been an outlet through which most of the material, now 
missing, was carried. But the present bottom is a series of undrained hollows, 
and there is no semblance of a stream course in its minor feattires. It is evident 
that something else has modified the form, and the only agency that could 
have done it is the wind. Here in the basin immediately north of Mount 
Uskvik, as in many other places, the wind is eroding and modifying in a minor 
way the sculpture work of former streams. There is thxis a superposition of 
topographies by different successive agencies."' 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History 0/ 
Central Asia, Amer. Mus. Nat. Hist., II, v-xxxi, pp. 1-475. 



We circled the lake and headed for a pass at the foot of Mount Uskuk, 
which stands more than fifteen hundred feet above the salt basin. Tucked 
away in a fold of the hUls were three yurts. While we were still more than a 
mile away, Shackelford shot at a flock of sand-grouse, and immediately we saw 
the Mongols, on horseback and afoot, fleeing into the ravines lilvc frightened 
rabbits. This happened not once but a score of times during the summer. It is 
grim evidence of what terror has been instilled into these peaceful dwellers of 
the desert by the tragic events of the last two years. At half a dozen places 
we heard the most gruesome stories of wholesale murder, and a rifle-shot or the 
sight of a single motor was enough to send an entire village into the hills. 

When we gained the summit of the first ridge and looked down into a gray- 
green desolation of ravines and gullies. Granger and the geologists agreed that 
these were Tertiary sediments and that evidently we were coming into a 
"badland" basin of considerable extent. 

As we looked over the coimtry to the south, Colgate and I wondered how 
it would be possible to get the motors through these ' 'badlands." It seemed as 
if nothing on wheels could go down the almost perpendicvdar sides of the ravines 
and get out again. The old Mongol said that there was a pony trail close to 
the side of Motint Uskxik, and while the geologists were away I prospected it. 
The guide was quite correct in saying that it was a "pony trail," for we can all 
swear that it was never meant to be traversed by anything else. During the 
next foiu" hours we took the cars into seemingly impossible places. Walls, 
rocks, ravines, washouts and ditches followed one another in rapid succession. 
Only Colgate's good driving got us through without a smash. 


At last we crossed the pass and were confronted with a broad river-bed of 
loose sand. This ended our troubles, and we camped at a well of sweet water, 
called Ondai Sair, only a mile on the other side. A yurt stood on a broad shelf, 
beside the well, surrounded by a flock of sheep which grazed eagerly on the fresh 
green grass where the ground was watered by the overflow. All the rest of the 
plain and the encircling hills vv'ere desert -like in the extreme — a floor of fine, 
sandy gravel scantily carpeted with sagebrush and hard bunch grass. 

Directly to the south of the valley, Baga Bogdo stands like a huge ram- 
part. Bathed in the violet light of sunset, it seemed like a fairy moiontain. 
We were all in high spirits, for it was evident that we stood at the entrance to 
a great sedimentary basin, splendidly exposed. 

The Mongols at the well said that they had come that day from the desert 
to the south. In the morning, when they had gone to collect their ponies, a 
wild ass was with the animals. That also was good news, for I was particularly 


anxious to obtain a group of wild asses for the Hall of Asiatic Life in the Ameri- 
can Museum. The Mongolian species was but imperfectly known, and no 
museum in America possessed specimens of it. 

The following day the geologists, with Granger, Colgate, Larsen and the 
guide, took a car down the valley to the spot from which the Mongols had 
reported fossils. At half past seven in the evening the exploring party returned. 
They reported the way down the valley to be very bad indeed, but possible for 
the cars. They had found a rather dirty well and a fine spring, and, in a three- 
hour search, discovered a few bone fragments. These were not very impressive 
but were sufficient to show that the strata were fossil-bearing. The country, 
they said, was like our western badlands in America — bare sedimentary hills 
and plains much opened by ravines and gullies and dotted with sculptured 
buttcs. Moreover, through the glasses. Granger had seen a wild ass com- 
fortably switching flies as he drowsed in the sun. Altogether it was distinctly 
encouraging. Because of the bad going down the valley, we decided to stay 
where we were until Granger and the geologists could give the surrounding 
covmtry a rapid survey from the palaeontological standpoint. 


It was necessary, however, to remain longer than we intended, for on 
Friday, June 23, winter fought a desperate rear-guard action with summer 
before accepting final defeat. Rain, hail and snow, with a temperature below 
freezing, kept us in oiu- tents, but two days later a smiling sun sent us happily 
on our separate ways into the field. 

Granger returned about sunset with no mammalian fossils, but he reported 
a considerable deposit of insects, crustaceans and small fish, in "paper-shales." 

The geologists did not come in until nine o'clock. They said they had 
found fossils but would tell little until after dinner. Then their spoils were 
spread out upon the table. They had many fragmentary bones, but these 
were so broken that it was impossible to identify them. Nevertheless, Granger 
felt sure that one portion of a leg-bone which Doctor Berkey had found must 
be dinosaurian. If that were true, it meant that the formation from which 
they had come was probably Cretaceous. As the geologists had seen no Cre- 
taceous beds since leaving Iren Dabasu, this would be an important discovery. 

We remained far into the night, examining the specimens and discussing 
the possibilities which had been opened by the day's work. These were the 
hours which I enjoyed most of all, and they will remain vividly in my mind 
when other memories have faded with the passing years. The fish, insects, 
Crustacea and plants proved to be most interesting and important. Some of 
the forms were new, but a fauna which is in part identical with it has been 
reported from eastern Siberia. The geologists believe that these paper-shales 


were formed in sheltered ponds of such quiet water that insects which died 
upon the surface sank to the bottom and were gently covered with a blanket of 
the finest sediment. In this matrix their tiny bodies left perfect impressions 
as the animal matter decomposed. The shales could be separated into sheets 
as thin as paper, and some of the specimens were so beautifully preserved that 
the delicate wings of giant May-flies, Ephemeropsis, were almost perfect in 
detail. One wing was 35 millimeters in length. 

The small fossil fishes, Lycoptera middendorffi, were exceedingly abundant 
in the paper-shales. They represented a new family, Lycopteridae, which is 
apparently ancestral to the Cyprinidae, according to Professor T. D. A. Cock- 
erelP who identified them. 

A dinosaur skeleton, of which I will speak later, was subsequently found 
by Granger in this region. The formation was designated as the Ondai Sair 
and has been ascribed to the Lower Cretaceous. 


The camp proved to be just as interesting zoologically as from the geo- 
logical and palaeontological standpoints. My traps yielded a rich variety of 
small mammals. Most of the species proved to be new to my collection and 
several of them new to science. In addition to marmots, spermophiles, kan- 
garoo rats, hamsters and voles, I caught specimens of Lagurus przewahkii, a 
pale yellowish mouse with very small ears, minute tail, and feet with heavy 
soles, which I had never seen before. It was originally described from Tibet, 
and its range thus is extended some eight hundred miles to the northeastward. 
Another new animal was a very beautiful jerboa which Dr. G. M. Allen, 1925, 
has named Stylodiptis andrewsi. 

A great slide of broken sandstone behind camp was inhabited by niunbers 
of picas or conies, Ochotona pallasi. One of my traps held a full-grown male, 
alive but with its back broken. I took the little fellow to camp, and although 
the hindquarters were paralyzed it appeared to be in no pain and ate grass- 
roots greedily; neither did it exhibit the slightest fear. This was due, pre- 
sumably, to the nerve shock which it had sustained. 

We also fovmd hares fairly abundant. Dr. G. M. Allen, 1927, considers 
this species to represent Lepus tolai tolai, which has a wide range throughout 
the Gobi Desert. 

It was interesting to encounter marmots again, for we had seen none 
since leaving the grasslands. Their presence was due to the proximity of the 
Altai Mountains. We discovered later that the typical fauna of the far western 

' Cockerell, T. D. A. 1924. "Fossils in the Ondai Sair Formation, Mongolia," Bull. Amer. Mus. Nat. 
Hist., LI, pp. 129-144; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 37, pp. 
129-144, 1926. 


Altai had followed the long finger of mountains eastward as it reaches into the 
desert, thus giving a much extended distribution not only to mammals, but to 
several species of birds. 


It was so apparent that we were in an extremely important region that 
Berkey and Morris decided to make a geological and topographical map, from 
the salt lake straight south down the valley, as a tj-pe section of Mongolian 
geology. It was a colossal undertaking with the limited time at their disposal. 
Since they had no other assistants and could not use a motor in that rough 
cotmtry as they had done at the Hot Springs, we hired two diminutive ponies 
from a Mongol lama whose yiut was five or six miles from camp. 

Granger thought it wise to go with us to the well at the southern end of 
the valley, where I wanted to hvmt wild ass. If the geologists found important 
fossil beds in their survey he could return for a thorough exploration. 


On June 26, we left the geologists wath one car, a chauffeur, a cook and a 
Mongol camp assistant. The rest of us turned southward toward the well ten 
miles away. The going was not as bad as we had expected, for in the lower 
part of the valley the sandy river-bed had become so solidly packed that the 
cars could run upon it as upon a floor. Then we climbed to a gravel plain 
which slopes gently southward from the base of a range of low lava-covered 
moimtains to the foot of Baga Bogdo thirty miles away. 

The tents were pitched above and to the west of the well and spring. 
This camp we later named "Wild Ass Camp." The country was desolate but 
strangely fascinating. From the door of my tent I could look south to the 
splendid mass of Baga Bogdo, its snow-capped peaks whiter than the clouds 
which always drifted about their summits. In the distant foreground was a 
long, flat-topped ridge, blood-red except for an upper gray- white stratum. 
Nearer were other hills and buttes of red, white and yellow, cut and sctdptured 
by the winds and rain. To the west the gravel plain, sparsely studded with 
desert vegetation, stretched away to meet the black mass of a lava flow; east 
of us across the wide river valley a similar gravel plain, extending far beyond 
the range of human sight, lost itself in the ever-changing mirage. 

Summer had come in a day and breathless stUlness lay upon the desola- 
tion of painted badlands. The flowing waves of heat gave weird fantastic 
shapes to rocks and grass; antelopes seemed to dance on air and flying birds 
to nm upon the ground. Lakes with reedy shores and wooded islets appeared 
where we knew there were no lakes, and somber forests offered the coolness 
of shaded glens. It was an tmreal world, menacing yet alluring. 


As I was gazing across the plains, the sky darkened, and a subdued roar 
came out of the north. I felt a sudden blast of cold air and turned to see a 
storm sweep from the river valley and rush away to the west at race-horse 
speed. In its wake lay a narrow trail of white hailstones, as large as pebbles. 
A moment later the desert was flooded with yellow sunlight which seemed to 
have passed through amber glass before it reached the plain. Throughout 
the summer this narrow track over which the hailstones were spread remained 
as a well-defined band of green. Thus quickly does the desert respond to the 
slightest moisture. 

Larsen was standing beside me, watching the rush of the storm through 
field-glasses. Suddenly he gave an exclamation and pointed to a cloud of dust 
less than a mile away. In the midst of it we could see three dun-colored ani- 
mals. They were wild asses! One was standing quietly while a huge stallion 
chased the other in circles. 

Five minutes later, Colgate, Larsen, Shackelford and I were in a car 
speeding toward them. We were all excited for it was our first sight of a new 
animal and one we had come far to seek. While we were still half a mile away 
they began to run west by south but going rather slowly, now and then stopping 
to glance back at us. They looked very neat and well-groomed in their short 
summer coats, and galloped easily. Suddenly they disappeared in a shallow 
draw with a narrow rocky entrance where it seemed we could not follow. 
Colgate jammed on the brakes, and by the time they were in sight on the 
opposite side we had opened fire, but they were beyond the range for accurate 
shooting and our bullets did no harm. They ran south from the valley into 
sandy ground where we could make no speed. Reluctantly we admitted 
that they had outgeneraled us. On the way back to camp we saw four more 
asses — two mares, each with a colt, — but they, too, kept to the sandy plain 
and left us far behind. The first chance had given us valuable experience, 
for we realized that it was useless to follow the animals when they were south 
of the gravel plain on which the camp was pitched. 

Twenty miles from our "Wild Ass Camp" was a large body of water called 
Tsagan Nor (White Lake). We could see its glistening siu-face distorted by 
the dancing mirage and the guide said that wild asses were very abimdant near 
its shores. The low ground, however, was so difficult for the cars to traverse 
that we did not reach the lake until two weeks later, when Shackelford and I 
discovered a practicable route. 


The day after our arrival at "Wild Ass Camp" I fotmd my first important 
fossil. We covdd prospect within a dozen yards of the tents, which were on 
the edge of a red and white wash. In the morning Shackelford found a 


beautifully preserved foot bone, of unidentified rhinocerotid type, in the 
bottom of a deep ravine. I was stimulated to better his discovery. After 
setting a line of traps in the river bottom near the well, I was wandering 
slowly along the sides of the ravine looking for fossils. Fragments were 
abundant but none was identifiable or worth keeping. Suddenly my eyes 
marked a peculiar discoloration in the olive-green upper strata; then I could 
see bits of white which looked like crumbled teeth. Scratching away the soft 
clay -like earth, I exposed the grinding surface of some large teeth and realized 
that it was something interesting. There was an almost irresistible tempta- 
tion to dig further and see what lay below, but I knew that if I did the wrath 
of the palccontologist would descend upon my head. The teeth were literally 
in powder, and fell apart in a thousand tiny fragments when the supporting 
earth was removed. Because of its bad state of preservation. Granger was 
doubtful at first whether it would be worth while removing the specimen, but 
eventually he decided to make the attempt. 

It took four days of intermittent work to get it out, and no one but a 
master of the technique like Walter Granger could have accomplished it at all. 
By means of a fine camel's-hair brush, he removed the sand almost grain by 
grain, wetting the teeth with gum arabic as each minute section was exposed, 
and stippling Japanese rice-paper into the crevices. When the gum and paper 
dried, the dust-like particles of enamel were so cemented that it was safe to 
expose a still larger surface. Then Granger soaked strips of burlap in flour 
paste and bandaged the fossil as though it was a broken limb; after a day of 
sun this formed a hard shell in which the specimen was safe. 

As the work progressed it became evident that much more than a set of 
teeth lay buried in the hill ; one side of the palate was exposed, then the jugal 
arch which forms the cheek, and finally the anterior part of the skull with a 
pair of extraordinarily long, decurved nasal bones. The teeth showed that the 
animal was a rhinoceros of a kind that none of us knew. Subsequently it was 
studied by Professor Henry Fairfield Osbom, 1924, who named it Baluchi- 
therium mongoliense. The formation was designated as the Loh, from the 
name of the region, and the age determined as Lower Miocene. It was in 
this same layer that Granger discovered the remains of a serrate-toothed 
mastodon, named by Professor Osbom Serridejitinus mongoliensis. 

My initial experience as a palseontological collector made me very keen. 
Every find was pregnant with the most interesting possibilities, for the veriest 
fragment of exposed bone might lead the way to a skull or skeleton ; moreover, 
each specimen turned one more page in the prehistory of central Asia. In 
every leisure moment I wandered over the badlands hunting for new treasures, 
but I was far behind Shackelford as an amateur collector. In some ixncanny 
way he seemed to know exactly where the best specimens lay, and he always 


came into camp with his hands or pockets full of teeth or bones which it seemed 
no one else could find. We were certain that he had developed an extra sense 
whereby he covdd smell an animal that had died millions of years ago. Granger 
was pleased at our efforts to discover fossils but his approval ceased abruptly 
when it came to removing them. My favorite tool was a pick-axe, while he 
used a camel 's-hair brush and a pointed instrument not much larger than a 
needle. When a valuable specimen had been discovered he would suggest that 
we go on a wild ass htmt, or anything that would take us as far as possible from 
the scene of his operations. 


In the red clays Granger found a small but highly important specimen. 
It was a beautiftilly preserved spadefoot toad. Dr. G. K. Noble, who described 
it and named it Macropelobates osborni, says it "is of unusual interest as repre- 
senting the group from which the modem spadefoot toads arose, to spread on 
one side across Europe and on the other into North America. It is the oldest- 
known fossil which belongs unquestionably to the Pelobatidas. . . . 

"The formation consists mostly of sandy clays. The terrain during Oligo- 
cene times was therefore similar to that to which modem spadefoot toads are 
restricted, except that it may have contained more clay and less sand. The 
climate was apparently semiarid."^ 


We very shortly discovered a well-marked trail two miles south of "Wild 
Ass Camp." It ran east and west, and subsequently we learned that it was a 
main caravan route from Kwerhwating to Kobdo. Although in many places 
it goes through very sandy country, still it has been beaten so hard by the camel 
traffic of centtu-ies that oxar cars could run upon it as well as on coimtry roads in 
America. Shackelford and I followed it one day ten miles to the westward. 
It led us into a vast dry river valley filled with deep red buttes, and knolls 
and sculptured pinnacles. To the west was the gently rolling plain, slashed by 
shallow gullies; to the north the low mountain range of the Uskuk block, 
capped with lava, the red sediments appearing like vivid wounds from beneath 
the dull black shell. We named this trail the "Grand Canyon." Subsequently 
it proved to be a rich fossil locality of the Oligocene Hsanda Gol formation. 

Eastward from camp the plain breaks up into a much dissected region 
which yielded a few valuable small fossHs but in which good things were 
rather rare. Except for a small deposit of Pliocene sediments at the foot 

' Noble, G. K. 1924. "A New Spadefoot Toad from the Oligocene of Mongolia, with a Summary of 
the Evolution of the Pelobatida;," Amer. Mus. Nomlatcs, No. 132, p. i; Preliminary Reports, Central Asiatic 
Expeditions, Amer. Mus. Nat. Hist., I, No. 27, pp. 1-15, 1926. 


of Baga Bogdo, which we named the Hung Kvireh formation, the three 
localities which I have mentioned were our most fruitfid palaeontological 

Fossils were very abundant in and on the badlands at the "Wild Ass 
Camp." All the members of the Expedition — Chinese, Mongol and foreign — 
became so interested in hunting bones that they spent every available moment 
wandering over the hills and into the ravines. My own work in zoology and 
the general direction of the Expedition kept me so occupied that I had little 
time for anything else, still I usually reserved the evening hours for a walk 
over the bone beds. Then the wind died out, leaving a solemn stillness over the 
desert, bathed in the golden light of sunset. Baga Bogdo assimied her evening 
dress of lavender, changing to violet as the twilight deepened; her rugged 
outlines were softened and clothed in an ethereal majesty; she seemed like a 
beautiful cloud that had settled to rest for a moment upon the earth. The 
mountain affected all of us strangely ; even the natives felt its spell and loved 
to sit gazing with field-glasses into the mysterious depths of its darkened 

Over the red and brown buttes and hillocks, thousands of rodent-jaws and 
skulls had weathered out. In places it seemed as though they had been sowed 
like grain with a lavish hand. Some of the smaller skulls were in soft con- 
cretions while others were almost clean. Teeth and skulls of a queer, gopher- 
like, giant bathyerid, Tsaganomys altaicus, a new genus and species, were 
abundant, as well as another new rodent, Cyclomylus lohensis. The late Dr. 
W. D. Matthew, who studied the collection of specimens, remarked: "It sug- 
gests the Asiatic ancestry of the family, although it cannot be considered as 
even approximately ancestral to the living genera."^ In our first summer's 
work here, eleven species of fossil rodents were obtained, of which seven repre- 
sent new genera. 

Remains of small carnivora were discovered in numbers. The new spe- 
cies, including two new genera, already have been described by Doctor Mat- 
thew, 1924. Insectivora numbered only fotu- new species and one new genus. 
The Perissodactyla are represented by two genera and species and the Artio- 
dactyla by one. The single representative of the Artiodactyla, Eumeryx 
cuhninis, is of particvdar importance because Doctor Matthew believes that 
it fulfils "more nearly than any hitherto described the required characters for 
an early Oligocene direct ancestor of the Cervidae."^ All seem to represent 

' Matthew, W. D., and Granger, Walter. 1923. "New Bathyergidas from the Oligocene of Mongolia," 
Amer. Mus. Novitates, No. lOi, p. 4; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. 
Hist., I, No. 19, pp. 1-5, 1926. 

2 Matthew, W. D., and Granger, Walter. 1924. "New Insectivores and Ruminants from the Tertiary' 
of MongoHa, with Remarks on the Correlation," Amer. Mus. Novitates, No. 105, p. 4; Preliminary Reports, 
Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 22, pp. 1-5, 1926. 


an Oligocene stage of evolution and the deposit of red and brown clays was 
named the Hsanda Gol formation. 


In closing a brief summary of this fauna Doctor Matthew, 1924, re- 
marked : 

"The character of the Baliichitherium fauna is peculiar as compared with 
most Tertiary mammal faunas, in the great abundance and variety of rodents 
and small camivora, and scarcity of ungulates, especially artiodactyla. It 
represents probably a somewhat different facies from the badland faiinas of 
western America, or the fissure and quarry faunas of western Europe. It may 
perhaps be a desert basin fauna. The association of true though primitive 
Cervidas with a fauna rather closely correlated with the older OUgocene of 
Europe and America is of importance as indicating the Asiatic origin of this 
group, if the principles of evolution and dispersal be adopted which were out- 
lined by Matthew in 'Climate and Evolution.' Hasty conclusions, however, 
are to be deprecated, as the evidence is still scanty and imperfectly studied and 
there is excellent prospect of obtaining more of it in the near future."^ 


During the weeks that we remained in this great Tsagan Nor basin, 
Berkey, Morris and Granger, 1923, made a study of it from the combined 
geological and palasontological standpoints. An abbreviated review of their 
conclusions follows, recording the section from the oldest to the yoimgest. 

Upon the oldrock floor, consisting of schists, marbles, graywackes and 
slates, all of pre-Cambrian age and invaded by granite, lie, near Mount Uskuk, 
the Ondai Sair sands and paper-shales of Lower Cretaceous age. These are at 
least five hiindred feet thick, and are faulted and tilted. It was in the paper 
shales that fish, insect and plant fossils were found. Resting unconformably 
upon the Cretaceous, lie about thirty-five hundred feet of early Tertiary gravels, 
sands and sandy clays. They include at least one lava flow. The higher beds 
are judged to be of Lower Oligocene age and are called the Hsanda Gol forma- 
tion. They carry the Baluchitherium fauna. 

A layer of clays less than one hundred feet thick is imposed upon the 
Hsanda Gol and is named the Loh formation. These upper beds are sparingly 
f ossiliferous ; the fauna indicates that they are Lower Miocene. 

Near the base of Baga Bogdo are later deposits of sands and clays, perhaps 
two thousand feet thick, judged to be of late Pliocene age, and designated the 
Hung Kureh. Of still later age is the mantle of coarse rubble, derived from 
Baga Bogdo and flanking the motmtain, which is at least two thousand feet 

1 op. cit., p. 6. 


thick and, so far as we know, is unfossUiferous. It may be very late Pliocene 
or Pleistocene. 

The thickness of the exposed sediments in the basin is between eighty-five 
htmdred and ten thousand feet. Only a fraction of these strata was acciimu- 
lated in any one period. This basin, which contains the oldest and the young- 
est basin sediments of which we have record, is the longest-lived and most 
active basin we have yet observed in Mongolia. 



The Tsagan Nor basin was exceptionally well suited to our purposes, for 
it furnished work for all the members of the Expedition. Berkey and Morris 
had a splendid type section of Mongolian geology to study and map ; Granger 
and his assistants were busy every moment with an abtmdance of new fossils; 
Shackelford reaped a harvest of motion and still photographs, and my hands 
were more than ftdl with the zoology. 

I had a splendid opportvinity to study the wild ass, a mammal new to me 
and but little known to the scientific world. No museum in America had 
specimens of the Mongolian species, and I believe there were only one or two 
skins in England, although the Tibetan species, Equus hemionus kiang, was 
fairly well known. J. H. Miller has given a short account of the wild ass of 
Dzungaria,' and a photograph of a dead specimen as well as of a yoimg living 
animal which had been domesticated by a Chanto native. The afternoon of 
our arrival at Loh we saw three wild asses, as I have already described, and I 
got the first specimen two days later. Just after tiffin, Larsen discovered a 
fine stallion not more than a mile and a half from camp. He was drowsing in 
the sun and stood absolutely motionless except for an occasional flick of his 
tail and lazy movements of his long ears. We watched him through field-glasses 
for a time and then Larsen, Colgate and I started out in the car. 

Profiting by previous experience, we ran almost due south and cut him 
off from the sandy ground on the lower plain. He seemed to divine otir inten- 
tions at once and ran for all he was worth toward Baga Bogdo. Colgate 
stepped on the gas and the motor leaped forward at forty-five miles an hour. 
The animal could not do better than forty miles, even when straining every 
muscle to cross in front of us, and that extra five miles was just enough to decide 
the race. He turned back on the plain and headed straight for the black lava 
flow a mile to the west. 

' Carruthers, Douglas. 1914. "Unknown Mongolia," with three chapters on sport, by J. H. Miller. 
Hutchinson & Co., London. II, pp. 602-608. 



It was thrilling when we rushed along within fifty yards of the splendid 
animal, the first we had seen in action at close quarters. I hated to give the 
word to stop, but he was dangerously near to the lava, and Colgate jammed on 
the brakes just as he crossed in front of us. The first volley turned him north- 
ward and we leaped back into the car to follow. He had a start of four or five 
hundred yards but was going perceptibly slower. To add to the excitement, a 
second ass appeared, seemingly from the air, and galloped parallel with us. It 
was almost within range, but the first one evidently was in difficulty and I 
fired again at three hundred yards. He winced, ran a few steps and rolled over, 
legs waving wildly in the air. We all yelled as he went down. It had been a 
great race and a new animal had been added to my long list of Asiatic game. 

I could hardly wait to examine the specimen. It was a stallion and proved 
to be the handsomest one we ever killed. The yellow-fawn color of the upper 
parts shades exquisitely into the pure white of the belly and rump patch. 
The mane is dark brown and short; from the mane a chocolate-colored band, 
margined with white, runs to the tail which is tufted and rather mule-like. 
The ears are longer than those of a Mongol pony but are by no means as large 
as a donkey's; in fact, the appearance of the animal is like a fine-bodied mule. 

In regard to the differences between the Mongolian wild ass and the 
Tibetan kiang, Miller remarks: 

"On comparing one of my specimens with an adult of the Tibetan kiang, 
Equus hemionus kiang, in summer coat, in possession of the British Musevun, 
I found several marked differences, as might be expected from the different 
environment of the two animals. The kiang is never found below an altitude 
of 15,000 feet; while the kulon of Mongolia rarely reaches an altitude of 
3000 feet, and, at any rate in Dzungaria and portions of Russian Turkestan, 
is found at an altitude of only 700 feet above sea-level. This accounts for the 
former carrying a much rougher and more wavy coat than the latter. The 
general body-color of a kiang is 'rufus' chestnut instead of the pale fawn of 
the more desert-loving kulon; its dorsal stripe is also less pronounced and 
without any light margin. But the greatest difference of all is that, in the kiang 
the whole of the legs are white, while in the kulon they are light sandy fawn, 
right down to the hoof. 

"My specimens are undoubtedly Equus hemionus typicus, called by 
Mongols 'Chigetai' and by Tiirki people 'kulon.' Its extreme eastern distribu- 
tion is at present imperfectly known ; Sir Francis Younghusband, in his journey 
across the northern Gobi, mentions seeing kulon in the Gobi at the extreme 
eastern end of the Altai. They are foiind north of the Altai Range on the plains, 
round the large lakes in the Kobdo region; we met with them near Barkul, and 
in several other places throughout Southern Dzungaria. Westwards they 
extend throughout Northern Russian Turkestan, being exceedingly numerous 


in the neighborhood of Lake Balkash. During the summer they frequent foot- 
hills, where the grass does not get so burnt up as on the plains; during the win- 
ter they roam all over the steppes, eating snow in place of water. The natives 
hunt them occasionally for their skins and meat, which they consider more 
palatable than the best mutton."^ 


In the days that followed the killing of our first wild ass, we hunted the 
animals industriously and soon had obtained all the specimens that were 
needed for musetmi purposes. Then we turned our attention to getting still and 
motion pictures, for this species had never been photographed. 

The Mongols told us the most amazing tales of the speed and endiu^ance 
of the "kulon," and I was anxious to obtain accurate data. One day we dis- 
covered a fine stallion well up on the plain near camp. It was in such a position 
that we could prevent it from reaching either the soft ground to the south or 
the western lava flow. Shackelford and I spent two hours in the car following 
the animal, besides exposing one thousand feet of motion-picture film and 
many stiU negatives. The ass was very clever in its attempts to reach the 
sandy grotuid of the lower plain, but each time we were able to turn it back. 

The highest speed that it could reach was forty miles an hour; however, 
this could be maintained for only a short dash, perhaps two fiirlongs. Sub- 
sequently we found that only a few of the fleetest individuals could reach that 
speed but that aU could do thirty-six miles an hour when galloping fiill out. 
Thus there was a difference of four miles an hovu* between the speed of the 
slowest and the fastest animals. 

To me the most amazing exhibit was the endtarance of the wild ass. The 
stallion which we followed traveled twenty-nine miles before it gave up and 
lay down. The first sixteen miles were covered at an average speed of thirty 
miles an honr, as well as coiold be estimated. During that time there never 
was a breathing space; it would sometimes slow up to twenty-five mUes an 
hoiu", when it had evaded us by a sharp turn, but a few moments later would 
speed up to a rate of forty mUes as it tried to cross in front of the car. Once 
we poimded along fifty feet apart for a considerable distance at thirty-six miles 
an hoiir. 

After sixteen miles the ass began to slow up perceptibly but kept doggedly 
at it, averaging almost twenty miles an hour for four miles more. Then he 
reduced his speed to a slow canter and resorted to more frequent twisting and 
turning to throw us off his track. Finally the animal stood quietly and 
Shackelford decided to lasso him. Fortunately, I did not fasten the end of 
the lariat to the car as Shackelford had suggested, for the instant the rope 

' op. cit., II, pp. 607-608. 


settled over the animal's neck it lashed out with both hindfeet, badly damaging 
the radiator, and then started of? on a sharp angle at a twenty-five mile an hour 

We had other opportunities later to check our observations as to speed 
and feel sure that we are correct in saying that forty miles an hour for a short 
dash is the greatest speed any of the Mongolian wild asses can reach. How- 
ever, this is considerably better than a wolf can do, for after several nms we 
were convinced that thirty-six miles an hour is the Mongolian wolf's fastest 
pace. The wolf is the only natural enemy of the wild ass. 

I was greatly interested in the way the wild ass runs. The head is held 
high and the tail low; the animal is always "collected" like a polo pony and 
never extends itself as does a race horse, except when putting on its utmost 
speed in a short dash. Because it is always so well collected it can reach full 
speed within a very few yards. This, I imagine, is a certain protection from 
wolves, for their attack would come only as a sudden dash from a ravine or 
other cover. None of the wild asses which we killed exhibited any indications 
of having been attacked by wolves, although they often showed scars which 
were obviously from wounds acquired by fighting among themselves. I wit- 
nessed several such battles — the two combatants rising on their hind feet and 
striking and biting fiuiously at each other. 

AU the wild asses which we lolled were in splendid condition. There was 
a thin layer of bright yellow fat vmder the skin, yet when the animals had had 
a hard race the sweat was clear, with none of the lather which proclaims a horse 
to be "out of condition." 


Miller remarks: "During the summer they frequent foothills, where the 
grass does not get so burnt up as on the plains; during the winter they roam 
all over the steppes, eating snow in place of water."' This, of covu-se, refers to 
the wild asses of the far west and Dzungaria. In the central Gobi where we 
encovintered them, they were feeding upon the dry desert vegetation such as 
camel sage, low thorny bushes and stiff bimch grass. This is the same food 
which the desert gazelles and our camels preferred. We never saw them close 
to the foothills where there was abimdant grass, even during the driest and 
hottest months of summer; they were invariably to be found well out upon the 
desert. The only time I have ever seen them eating green vegetation was in 
1925, when we explored a narrow valley about one himdred miles in length 
parallel with the base of the eastern Altai Mountains, just west of Ikhe Bogdo. 
There was a great concentration of game in this valley, wild asses, both grass- 
land and desert gazelles, and hares. At first we were at a loss to account for 

' op. cit., p. 608. 






1" '^"Tf^^— mjjp— '°™i 





it, until we realized that they were aU feeding upon a species of wild alfalfa. 
About the edge of Tsagan Nor, where we camped for several weeks, there was 
much green grass; at the western end a small river, the Tatsin Gol, nms into 
the lake, and for a number of mUes along the banks there is excellent grazing. 
Yet, we never saw wild asses in this vicinity, although there were literally 
thousands of them upon the desert plain only a mile or two away. 

Miller's statement that the "kulon" came to various springs to drink and 
that they eat snow in winter is interesting as differing from our experiences. 
The water of Tsagan Nor was only slightly brackish, and our camels drank it 
eagerly, yet we never saw either wild ass or gazelle at the lake nor were their 
tracks to be found in the marginal mud. I was particularly careful to examine 
the shore-line for such indications of their presence. At a spring near our camp 
at Loh there were so many tracks of ponies that we could not distinguish 
whether or not wild asses came there, although the Mongols said that they did. 
In the "Grand Canyon," to the west of Loh, there were many of their tracks 
about the spring. 

South of the Altai we explored one of the most barren deserts that I have 
ever seen. For many miles there was no water except at a few wells along a 
caravan trail. In this desert we saw a few "kulon, " but not many. I attributed 
their scarcity to lack of food, for there was virtually no vegetation of any sort, 
rather than to absence of water. It is my belief that wild asses reqtiire but very 
little water; that, as in the case of other desert animals, the starch in the 
vegetation which they eat is converted by digestive processes into water which 
is sufficient for their needs. In the Gobi, during the stunmer, there is almost 
no rain and little dew; there is no way for the animals to get water in many 
parts of the desert where they are abundant. 

The "kulon" and the desert gazelle live under like conditions in the same 
regions and are frequently seen together. In fact, if we followed a herd of asses 
for a few miles in the motor car, gazelles almost invariably came from either 
side to join in the chase. I feel certain that the gazelles do not require water. 
In 1925 we found a "kulon" and a gazelle which appeared to be inseparable 
companions. Day after day we saw them in the same place grazing together, 
never more than a few yards apart. On the plain at Loh we usually had one 
or more wild asses in sight from the camp. With the aid of field-glasses I used 
to watch them throwing dust over themselves, grazing or drowsing peacefully 
in the sun. Once or twice I saw one lying down. 


The "kiolon" of the central Gobi begin to drop their young during the last 
week of June, and we saw newly-born young as late as July 10. On July 6, 
1922, Shackelford and I caught a baby wild ass which we kept as a pet for six 


weeks. The little fellow was not more than three days old and ran beside his 
mother in a stiflf-legged, uncertain manner that was most amusing. He could 
not go very fast and before long stopped. Shackelford roped him without 
difficulty and we took him back to camp. He looked like a small woolly donkey. 
In a short time "Buckshot," one of our Chinese assistants, taught him to drink 
tinned milk from a canteen, and the little chap appeared to be quite con- 
tented with his new diet. A few days later we purchased three goats from a 
Mongol, but they did not give sufficient milk to satisfy him. 

I hoped to be able to bring him back to America for the New York Zoologi- 
cal Society, but on August i8 he died. He had lived precariously upon tinned 
milk and goats' milk, and appeared to be strong although he increased very 
slowly in size. At the Artsa Bogdo camp in August we got a plentiful supply 
of cows' milk from the Mongols, but it appeared to be too rich for the baby ass 
and he died within a few days. 

I have never seen such a wild, untamable animal. Even though it was 
treated with the greatest kindness by all of the men, it never lost its fear except 
with "Buckshot," who fed and attended it constantly. Whenever anyone 
else would approach the animal, it dilated its eyes and kicked viciously. Even- 
tually it learned to follow "Buckshot" like a dog and would even enter the cook 
tent if he happened to be inside. At the end of six weeks, when it died, it 
appeared to be even more frightened of the rest of us than when it was first 
captured. I have had many young animals of different kinds as pets but never 
have I known one to be as vmtamable as this wild ass. 

Possibly this may have been due, to a certain extent, to the fact that the 
little fellow was a stallion. On June 24, 1925, McKenzie Young and I captured 
a female which was not more than a day old. This one appeared to be much 
less frightened and to have a more gentle disposition. Unfortunately, it 
escaped after two days so that we did not have an opporttmity to learn more of 
the comparative docility of the sex. As it was a cold evening the baby ass had 
been buttoned up in a leather waistcoat belonging to Lieutenant Robinson 
and had my police dog's collar about its neck. The rope became loosened dur- 
ing the night and the foal wandered off. Doubtless it died, for certainly no 
other ass would allow it to approach with the man scent upon the vest. 

If I were to try again to bring one back for the Zoological Society, I should 
purchase a pony mare from the natives and teach the little ass to nurse from 
her, for I am convinced that it never would prosper upon other milk. A 
Mongol told me that he had reared a wild ass colt which he persuaded a pony 
mare to adopt. At first the mare would not allow the youngster near her; 
then the Mongol conceived the idea of pouring some of her own milk over the 
baby ass and this soon solved the difficulty. The animal flourished but was 
always wild and difficult to approach, except by its owner. He never was able 


to break it to the saddle or use it in any other way. The ponies woiold not 
allow it in the herd with them. The Mongol bred it to a pony mare and the 
resulting mule, although a fine strong animal, could not be ridden. 

Mr. F. A. Larsen, who was with us in 1922, is a well-known horse dealer, 
and had hoped to capture several wild asses for breeding purposes, but after 
our experience and that of the Mongols, I believe it to be very doubtful whether 
they would produce usable mules. 

Like the desert gazelles, the "kulon" seeks a fiat plain upon which to drop 
its young. They are particularly careful to avoid a region of ravines or gullies 
which might give cover to wolves. Also like the gazelles, they gather into 
herds, largely composed of mares, just before the yotmg are bom. The stal- 
lions do not entirely leave them but remain somewhat separated. Later in 
the summer many of the males range by themselves. The solitary individuals 
which we saw were invariably stallions. 

The mares evince considerable affection for their yoimg and when we 
were following a colt in the car the mother would remain close to it, regulating 
her speed to that of the baby. It was interesting to see how quickly the colts 
were able to run. Several that we estimated were not more than a week or 
ten days old could reach a speed of twenty-five miles an hour and maintain it 
for a mile or two. They tired quickly, of course, but it was by no means easy 
to capttu-e them, for they could dodge and twist with the greatest agility even 
though too exhausted to keep up a steady pace. I shot several yoiing for a 
group in the American Museum of Natural History, but had great difficiilty 
in getting good specimens. The colts were usually badly scarred, as if from 
bites, and two were suffering from a skin disease like eczema. 

The first two weeks of life must be the most critical for them, for then 
wolves could catch them easily if they were unprotected by their mothers. 
On the great plain north of Tsagan Nor, however, where two or three thousand 
wild asses had gathered to foal, I did not see a single wolf, and carcasses were 
left imtouched except by birds. A short distance to the west, across the Tatsin 
Gol, a small river empties into the Tsagan Nor, where a second great plain exists. 
It is exactly like the other, but there were few asses upon it and if they were 
driven away from Tsagan Nor they would drift back in a day or two. 


The "kulon" seemed to prefer the hard gravel plains and would nm into 
the sandy cotmtry only when we approached in the car. They appeared to 
know instinctively that we could not follow them there and would always 
make a direct line for soft ground, or into the lava flows which cap part of the 
region about Loh. 

The car had an irresistible attraction for them, as it did for the gazelles, 


ponies, camels and all other animals on the plains. When we were running an 
ass, it would make a supreme efifort to cut across in front of us, sometimes 
missing the car by only a few feet. If there were several hundred asses upon 
the plain, we needed only to drive up the center to have the animals come in 
diagonally from both sides, as though drawn by a magnet; after a few miles 
the whole mass would be thimdering along in front of us. Usually such a spec- 
tacle was too much for gazelles in the vicinity and they, too, would join the 

Miller remarks that the eastward range of the wild ass is unknown. We 
can give some accurate data upon the subject since we came always from the 
east. The most eastern record which I have is of a single individual that we 
encountered just south of the Gurbun Saikhan, of the eastern Altai, in 1925. 
This is latitude N. 43° 30', longitude E. 105°, and is six hundred mUes from 
Kalgan. In 1923 we saw a single wild ass near Artsa Bogdo, sixty miles farther 
west. Thirty or forty miles still farther west they became fairly abundant, 
and at Tsagan Nor hundreds can usually be foimd. 

As we explored the country for three seasons very carefully, I feel con- 
vinced that the Gurbun Saikhan is their easternmost range north of the Altai. 
South of the Gurbun Saikhan we saw none, although we traveled a long way to 
the east. Just why the wild ass should so suddenly stop at this point is rather 
puzzling, for the country to the east for several hundreds of miles is essentially 
like that of the Tsagan Nor region, with similar climate, vegetation and ter- 
rain. Westward, both north and south of the Altai, we saw wild asses, but in 
no place were they as abimdant as on the Tsagan Nor plains. 

MiUer says that the natives of the west consider that the flesh of the wild 
ass is "more palatable than the best mutton,"' but none of the Mongols whom 
we saw would eat it. They never hunted the animals nor paid any attention 
to them except to race one now and then with their ponies for sport. 

In Chapter XXVI, I have given an account of a great herd of wild asses 
which we found in 1925 at Tsagan Nor. 

' Loc. cil., p. 608. 



During the time that we camped at Loh, my small mammal traps yielded 
an abundance of specimens, which, however, represented a typical desert fauna 
much like that of Uskuk. To oiir surprise, we collected at Loh the first insec- 
tivore that we had found in the Gobi. This was a hedgehog, Erinaceus, which 
proved to be abundant on the shores of Tsagan Nor twenty miles away. 

Colgate and I poisoned the carcass of a wild ass which we killed on the 
plain west of camp. For one night it remained untouched, but on the second 
morning we foiind near-by two wolves as well as foxu* kites, one golden eagle 
and a huge black vulture, ^gypius monachus. The wolves still wore patches 
of their long winter fur even though it was Jvme 29. The comparative scarcity 
of wolves is shown by the fact that we did not get another, and did not even 
see one, although we poisoned carcasses in widely separated localities. Yet, 
this region was a favorite breeding-ground for both wild ass and gazelle, and 
there were hiondreds of young animals about. One would have supposed that 
wolves would congregate there from all the surrounding coimtry. As I have 
remarked before, I have seen more wolves along the Kalgan-Urga road than 
anywhere else in Mongolia, except at Wolf Camp in 1930. 


At every poisoned carcass we got several of the great black vultures. 
This huge bird is one of the most characteristic and interesting sights of the 
Gobi. One had a wingspread of nine feet six inches. The top of the head is 
not bare but is covered with short, downlike feathers. The feet are compara- 
tively weak, but the great hooked beak is a most effective weapon. They 
require space for a nm in order to lift themselves into the air, and I saw one 
which had so gorged itself upon the flesh of a wild ass that it could not get off 
the ground. When I approached, it settled back upon its tail, striking viciously 
with its beak. One day I shot a gazelle and left it to follow another wounded 
buck. When we returned to the dead animal half an hour later a black vulture 
flew away. We fovind that during that short time the bird had almost stripped 
the flesh from one side of the carcass. 



In 1925, Doctor Chaney took a young black vulture from a nest at Baga 
Bogdo and brought it to camp. It flourished and became as tame as a chicken. 
Eventually I took it back to the New York Zoological Park, where it still lives. 
From the very first we fed this bird upon fresh meat and it absolutely refused 
to eat carrion of any sort. If meat had the slightest decayed odor it would have 
nothing to do with it. Viscera seemed particularly distasteful to the vulture 
and only once or twice did we persuade it to eat a piece of antelope liver, when 
there was no other meat and the bird had had no food for thirty-six hours. 

The bird cared for itself in the most astonishing manner. If we were 
camped near a lake, it would wade into the water for a bath two or three times 
a day and then drowse in the sun with wings half spread while drying its 
feathers. It was always allowed the freedom of the camp and never attempted 
to get away. In fact, it got distinctly lonely if most of the men were gone and 
always preferred to be near someone. Its favorite sleeping-place was in the 
rear of my tent; my police dog also liked to sleep there and the contests for 
supremacy were most amusing. The dog was usually worsted in these encoun- 
ters, for he evidently considered it beneath his dignity to fight with a bird. 

If we were camped near a spot where there were cliffs, the vulture wovdd 
spend hours sitting on a projecting pinnacle gazing over the country below. If 
there were no cliffs, the bird seldom left camp. I was much surprised at the 
amount of water which it consumed. Drinking by itself from a pail was too 
slow a process. It much preferred to open its great beak, throw its head back 
and have someone pour water down its throat. It had considerable intelli- 
gence. One day I was sitting in my tent writing. A gasoline tin of drinking 
water was near the door. The vulture came up to the tin and rapped upon it 
with its beak, significantly. I paid no attention and after three or four raps 
the bird entered the tent, jerked my coat and returned to the tin. Of course I 
gave it water. I could hardly credit the performance but there was no mis- 
take; the vulture knew there was water in the tin and that it could not be 
had without human assistance. As a matter of fact, it had been given water 
very often from the tin, which was usually kept at the tent door. During the 
long trip from China to New York, the bird became very much attached to 
me and would recognize me instantly even when there were other men about. 
It was extraordinarily curious and when the men were packing fossils it insisted 
upon examining every box. Although Mongolia is its summer home, the black 
vulture goes southward during the winter, and stragglers have been reported 
even in Fvikien Province, near Foochow, south China. 


At Loh I observed an interesting habit of the golden eagle. We were 
driving across a perfectly flat plain with two cars. I saw a full-grown golden 


eagle crouching behind a small bush with its head stretched out. After we 
passed, it half rose to its feet, and then it saw the second car. It sank back 
again and remained motionless until the motor had passed, when it flew away. 
We observed this same trick by other individuals at two other times during 
the summer. 

The kites were a never-ending source of amusement to us at camp. Thirty 
or forty of them were usually sitting on the ground or flying about the tents. 
As soon as one would pick up something and attempt to fly away, the others 
would attack it like a pack of wolves. One day when there were a great 
number about we threw out a dozen bits of meat. The kites sat in a row a 
few yards away for more than an hour. Each time one made an attempt to 
get a bit of food the others flew at it. As a result none of them got any of 
the meat, although they all wanted it. 


The Fourth of July at Loh was unforgetable because of the mirage. It 
was the first day of real heat, the temperature rising to 95° F. in the shade, 
and absolutely windless. The desert swam in heat waves. If a man walked a 
hundred yards away from the tents he seemed to be wading in water up to 
his knees. Everything was so distorted that it seemed as if we were living in an 
unreal world. In the afternoon we were treated to another vmusual sight. 
The whole shimmering plain became alive with "wind devils," small whirling 
clouds which sucked the sand and dust into the air perhaps one hundred feet 
high and danced away at race-horse speed. One of them struck our camp 
squarely in the middle; socks, trousers and shirts which were drying on a 
line were drawn up into the air and scattered over the plain. The wind 
devil passed in thirty seconds but it nearly wrecked our tents. Once I 
raced a large whirlwind in the car and found that it traveled faster than 
thirty-five miles an hovir, which was the best speed I could make on the 
rather rough terrain. I estimated that it was moving at about forty-two 
miles an hour. 


On July 3, I sent Colgate, with Larsen and Badmajapoff, to Urga with one 
car. Larsen and Badmajapoff had to return to their business interests, and I 
was particularly anxious to get news of what was happening in China, for a 
sizable war had started just as we left. I disliked to send a single car on an 
eight-hundred-mile trip, but we knew that the traveling was good most of the 
way, and Colgate felt confident that he would have no difficulty. As a matter 
of fact, he made the trip in just twelve days, including four and a half days 
spent in Urga. 



Shackelford and I had made several attempts to reach Tsagan Nor, the 
lake which we could see gleaming whitely twenty miles to the south. Each 
time it had been impossible to go very far because of the soft sandy ground, 
but on July 6, we had better success. Much to our surprise we were able to run 
quite easily over the southern fiat, and then we realized that all our earlier 
attempts had been made a day or two after rain when the ground was still soft. 
When it had been baked by the sim there was sufficient crust to hold up a light 
car. Half-way to the lake we came upon another fine hard plain, which 
swarmed with wild asses and antelopes. At the western end of Tsagan Nor, 
where a small stream, the Tatsin Gol, empties into the lake, there was a wide 
valley carpeted with green grass. There fifteen yurts had been pitched along 
the base of the bordering sandy hillocks, and the Mongol inhabitants rushed 
out to wave us a greeting. They were not frightened of the cars, as they had 
known of our presence since the first day that we arrived at Wild Ass Camp. 

The lake proved to be a lovely spot. Coarse green grass margined the 
water, and Baga Bogdo was reflected in its calm surface. Himdreds of ducks, 
geese, sheldrakes and grebes were paddling about, followed by trailing wakes 
of downy young. Between the lake and the mountain south of it there was a 
long line of cream -white sand-dunes, beautifully sculptvu^ed by the wind. 

We were both so enchanted by the lake that we decided to move our part 
of the camp there at once. The geologists had arrived from Uskuk, having 
reached Loh in the southward extension of their geological and topographical 
map. They had foimd the region to be even more interesting from a structural 
standpoint than they had at first supposed and were keen to carry their work 
over to Baga Bogdo. Granger thought it best to remain at Ivoh for further 
palaeontological work. After I had made a trip back to the caravan for sup- 
plies and instructed them to join us at the lake, the rest of us were free to 

Shackelford and I went down on July 1 1 and camped on the gravel beach 
not fifty yards from the water's edge. Short grass and weeds gave it quite a 
lawn-like effect. Neither of us has ever forgotten that first evening. Just as 
the sim disappeared, Baga Bogdo was flooded with a wonderful lavender light 
which edged the lake with deepest purple; then the moon rose from behind 
the sand-dunes in a splendor of gold, drawing a glittering path across the water 
to the very door of our tent. 


In the marsh-grass and rank vegetation beside the lake, a green insect, 
like a large mosquito, swarmed in countless thousands. Just at dark these 
began to rise with a hum like distant motors. The noise was quite appalling 



and we thought that we should be forced to leave, but forttinately the insect 
is exclusively a vegetable feeder and did not annoy us in the slightest. They 
formed a stratum three feet thick and seemed to be following the lake shore 
from west to east. The flight line lay four feet above the groimd, and below that 
level there was hardly an insect. Not many came into the tents when the 
candles were lighted, and there were no mosquitoes or sand flies; in fact, it 
was an ideal summer resort. 

Of course we were curious to know whether or not there were fish in the 
lake. The first evening we saw a number of suspicious-looking swirls in the 
water which we were certain must be made by fish. Hooks and lines gave us 
nothing, but I had a twenty-foot seine which yielded a plentiful supply of 
minnows. They were all of a single species, Oreoleuciscus pewzowi, and few of 
them were more than eight or ten inches in length. Mr. J. T. Nichols, 1930, 
who has identified the fish, says that their affinities are to species of the far 
western Altai Mountains. How these small inland lakes became stocked with 
fish remains a mystery. The supposition that the eggs or the fish themselves 
were carried from one lake to another by birds does not appear entirely satis- 
factory in a region where the bodies of water are so widely separated as in 
the Gobi. There are gulls, terns, ducks, geese and sheldrakes on almost all of 
the lakes, but in Orok Nor and Kholobolchi Nor, two lakes only thirty miles 
from Tsagan Nor, the fish of the first two named lakes were unlike those of the 
last. It seems more probable that at some time in the past there has been a 
drainage into the lake of small streams which have disappeared. As we know 
that some, at least, of the smaller lakes which now contain fish have been dry 
for one or two years during certain intervals, it is evident that the fish must be 
capable of living buried in the mud for long periods. The problem is one 
which requires more study. 

At Tsagan Nor, and in fact all through the desert, two species of lizards 
were abundant. These have been identified by Mr. Karl P. Schmidt, 1927, 
as Phrynocephal-HS versicolor and Eremias przewalskii, the former ranging south 
into northern Shansi and being much more numerous than the latter. Except 
for these two species we saw no lizards in the Gobi. 

In the coarse grass at the western end of Tsagan Nor, where the Tatsin 
Gol flowed into the lake, mallard ducks, ruddy sheldrakes and bar-headed 
geese had nested, and their yotxng were about one-third grown. The birds 
soon became accustomed to our presence, since we did no shooting near camp, 
and would bring their broods close into shore when the wind blew the feed 
toward our side of the lake. Shackelford got many photographs of the birds 
from his tent. 

One day in late July I discerned a pair of geese which I coiild not identify, 
and shot one. It proved to be the graylag, Anser anser, and a few weeks later 


we saw several flocks of the same species some miles to the eastward. I have 
not seen this goose at other localities in Mongolia and conclude that it is not 
abundant there. In north China it is only an occasional visitor. 

Our traps yielded many small mammals, among which the two sand rats, 
Meriones aticeps and M. imguiculatus, were in greatest abimdance. Although 
the two species are much aUke in appearance and live side by side, the former 
is strictly nocturnal while the latter is almost entirely diurnal. In a dry 
stream-bed at the "Grand Canyon," ten miles north of the lake, Granger 
discovered a colony of giant sand rats which proved to represent RJiombomys 
opimus nigrescens. We had previously obtained a single specimen at Iren 
Dabasu, which extends the range well to the east. The animals live in colonies 
like the smaller gerbils, burrowing in the sand hummocks. This species appears 
to be cntirel}' diurnal. 

One night at Tsagan Nor, a very small brown shrew, the only one we saw 
in the desert, was caught in the taxidermists' tent. Dr. G. M. Allen, 1928, 
has described it as a new species, Crocidiira lar, and remarks that the genus is 
thus extended in its northward range in this part of Asia. It certainly made a 
mistake in choosing the taxidermists' tent for its investigations! 

The hedgehog was the only other insectivore that we caught in the Gobi. 
These animals were abundant in the long grass near the water's edge. We 
would throw a piece of antelope meat into the grass and shortly after dark 
visit it with a flashlight. Usually we could discover one or more hedgehogs 
in the vicinity and could catch them without much difficiilty. The older 
individuals were not easily tamed, but one half -grown male became as gentle 
as a kitten. In fact, Mr. Shackelford kept this little fellow as a pet all through 
the summer and eventually took it to New York with him. After living for 
some time in his apartment it was transferred to the New York Zoological 
Park. It ate grasshoppers and other insects and bits of fresh meat voraciously 
and was particularly fond of big black beetles. After a few days it could be 
handled with impimity. Only when angry or suddenly startled woiild it 
erect its quills. Surprisingly enough, it seemed to require a good deal of water 
and wotdd lap it up from a shallow plate like a dog. When we returned to 
Peking in the autumn, by mistake one night the hedgehog was placed in a box 
containing a young and active alligator. Alligator sifiense, about eighteen inches 
long. In the morning the alligator was dead. The hedgehog had almost 
entirely devoured the right hind leg and had eaten a large hole into the abdom- 
inal cavity. 

One day at Tsagan Nor I shot a fox, Vulpes vulpes karagan, a pale straw- 
yellow animal such as one would expect to find in the desert. I was stalking a 
flock of ducks, which were swimming near the reed-grown shore, and discovered 
the fox engaged in the same pvirsuit. 

M _M 






XOH, 1925. 


Procapia gidlurosa. 


Cazclla siihgultiirosa hillicriaiia. 




On the gravel plain north of the lake, gazelles, Gazella subgutturosa hillier- 
iana, were very abundant. The great flat plain was an ideal place for the 
does to rear their young. When we reached there on July 1 1 , most of the baby 
antelope were two or three weeks old and were running with their mothers. 
The does begin to drop their young about June 15; we saw no newly-born 
fawns after July 10. Virtually all the does were in milk. A single young is 
usual, but twins are not infrequent. 

The grassland gazelle, Procapra gutturosa, gathers into great herds of 
does in the spring just before the young are bom and again into mixed herds 
in the autumn. The desert gazelle never does this. I believe the reason is 
that at no spot in the desert is there sufficient vegetation to support a large, 
slowly moving herd; in the grasslands there is ample feed. 

On the Tsagan Nor plain the gazelles were in groups of from half a dozen 
to thirty or forty, all bucks; the females usually were alone or in couples, 
each with its fawn. Only once or twice during the summer did we see bucks 
and does together. The fawns were very clever at hiding. They would lie 
flat upon the groimd beside a sage bush only a few inches high, with their ears 
dropped and neck stretched out. Many times I have tried to creep up and 
throw a coat over one, but just before I was within reach it would dash away. 
The brown woolly hair was so exactly the color of the desert gravel that the 
little fellows were almost invisible when motionless. I usually discovered them 
by their brilliant eyes. 

The fawns hide until they are about two weeks old; then they prefer to 
trust to their legs and speed. Nevertheless, they can run at a very respectable 
rate within a few hours of birth. One which was not yet dry behind the ears 
tired out my pony in a chase; although the fawn could not run as fast as the 
pony, by rapid turns it would gain so much distance that I could not catch it. 

We followed a fawn on June 27 in the motor car. It was about ten days 
old and was with its mother. When we first discovered them, the female tried 
to entice us away, but we stuck close upon the heels of the fawn. For the first 
four miles it averaged twenty-five an hour, but soon tired after that and 
resorted to dodging to elude us; it ran a total of nine miles before we finally 
caught it. After the first three weeks a wolf could not equal the speed of a 
baby antelope. We did not find any wolves that could exceed thirty-six miles 
an hour even on the first dash, and the fawns could reach forty miles an hover 
without half trying. 

We have demonstrated beyond a doubt, with otir cars, that the desert 
gazelle can reach a speed of sixty miles an hour in its initial dash. It can 
maintain this speed for only about half a mile; then it will drop to forty or 
forty-five miles an hour. The grassland species, which is larger and has a 


heavier body and proportionately shorter legs, cannot reach a greater speed 
than forty-five or fifty miles an hour. We followed one desert gazelle, a fine 
buck, for ten miles. We were on a flat plain where the going was excellent for 
the car. He left us easily in the first three miles and we could just keep his 
bobbing white rump patch in sight; then he settled down to a steady pace of 
thirty miles an hour, keeping about one hundred yards in front of the car. He 
continued at this speed for seven miles until we punctured a tire. During the 
last two miles his tongue was hanging out, but I am sure that he could have 
run five or six miles more. 

When the gazelles first start off they often progress in a series of stiff- 
legged bounds, which make them appear to be on springs, but when they 
reach high speed they settle into a smooth even run and seldom bound. They 
will not go faster than is necessary to keep about two hundred yards in front of 
whatever is following them, and it is only when thoroughly frightened that they 
will extend themselves. They always try to cross in front of a pursuer. 

Of course, the ability to reach such a high speed in the initial dash is a pro- 
tection from wolves, their only important natural enemy. Wolves can hope to 
catch an antelope only by lying in wait behind some cover until it is near 
enough for a svirprise attack. The gazelles' safeguards are being able to "get 
away" like a flash and leap into high speed instantly. 

Eagles are the principal enemy of the baby gazelles, and they must get a 
good many adults too. About an eagle's nest at Ardyn Obo I fovmd the remains 
of at least twenty antelopes, some of them being full-grown bucks. Miller says 
that in the western Gobi, the Kirghiz, and occasionally the Chantos, hunt 
gazelles by means of trained golden eagles, but, in the regions which we visited, 
the Mongols did not indulge in this sport. 

Gazelles have amazing vitality and even when badly shot can continue to 
run. With a broken foreleg the animal can easily reach thirty-five miles an 
hour; a shattered hind leg slows it up considerably but it can still do fifteen 
miles an hour for some distance. Although the grassland species sometimes 
wanders into the edge of the desert, I never have seen the desert gazelle in the 
grasslands. Like camels and wild asses it prefers the dry, hard vegetation of 
the desert to the most succulent grass. Miller says that the grassland species 
is found all over northwestern IMongolia, north of the Great Altai, and south of 
the Tannu Ola ranges. He puts their extreme northwesterly limit as Kash- 
Agatch and remarks that in the summer they ascend right up into the moun- 
tains where he hunted Oins amnion. He mistakenly considers Russian Turk- 
estan and western Asia to be the centers of distribution of the desert gazelle. 



Immediately upon our arrival at Tsagan Nor we made friends with the 
Mongols who were encamped at the western end of the lake. They were 
cordial as usual and we saw an opporttinity for Shackelford to photograph a 
good film story of native life. The yurts were arranged in three groups along 
the gravel base of the lake depression, at the edge of a green meadow formed 
by springs from the Tatsin Gol. 

A lama about fifty years old appeared to be the head man of the village, 
and we were invited for tea in his yiu^t upon our first caU at the encampment. 
His religious vows had not prevented him from "taking unto himself a wife" 
and he had a thriving family. One of his daughters, an attractive girl of 
eighteen, shyly approached me holding up a bandaged hand for inspection. I 
found that a finger had been badly injured and was a mass of gangrene. Sub- 
sequently I treated the finger with such good results that the family became our 
firm friends and we could do as we pleased in the village. 


The lama was well-to-do and owned large herds of sheep, goats and 
camels. Nevertheless, his yurt varied in few particulars from those of the 
other less opvdent members of the community. Several red chests, and beside 
them a small altar bearing a picture of Buddha, stood at the back; on the right 
side was a low bed platform six inches off the ground; near the door, on the 
right, a rack of milk-pans and several old skins upon which half a dozen very 
young lambs and goats were lying. The seat of honor is at the back of the yiu-t, 
facing the door. Yurts vary little in their interior arrangements, but some of 
the richer men keep a special reception yurt, in which young animals are not 


In the center of every yurt an argiol fire burns, and over it the family 
cooking is done in a great iron bowl. Tea and milk are boiled together, ladled 



out into tall cylindrical brass-bovind pitchers and passed about to be poured 
into the small wooden eating bowl which every Mongol carries in his, or her, 
gown. As often as it is available, mutton or other meat is boiled in the great 
pot, and eaten lAath the fingers and the aid of a sheath knife. 

The Mongols of the desert appear to keep cattle only for their milk, or to 
sell. I have not happened to see one kill a steer for, but I am told that 
they do eat beef to a large extent in the grasslands where cattle are more numer- 
ous. Mutton is their favorite food. They prefer it to any other kind of meat. 
Our caravan Mongols, who of course were somewhat spoiled, refused to eat 
antelope because they thought that by refusing antelope, they would get more 
mutton. They stoutly maintained that they did not like antelope meat; but 
other natives were delighted to get it. The mutton is good except where the 
sheep have been eating the small wild onion which grows abundantly every- 
where on the desert. Then the flavor is decidedly "oniony," but the Mongols 
consider that as a distinct asset. I have never seen the Mongols prepare meat 
in any other way than by boiling. Since they are a pastoral people they have 
almost nothing to eat except animal products. Butter, cheese, milk, meat 
and tea are their ordinary' food. Now and then they obtain a little flovu- from 
the Chinese traders and make a kind of bread fried in grease, but the poorer 
natives cannot afford to indulge in such luxuries. 

The Chahar and other IMongols who live in Inner Mongolia, where they 
are in closer touch with the Chinese and an agricxoltural region, use a great 
deal of flour and other vegetable foods such as potatoes and millet. On the 
slopes of the Altai Mountains we found a few wild onions and rhubarb. The 
latter the natives use only as a medicine, but the onions they eat raw. In the 
spring I have seen them eat the pods of young milkweed and in the autumn 
the fniit of the dune berry, Nitraria schoberi. 

One might suppose that such an exclusively animal diet would be unhealth- 
ful. Such is not the case. During the long, extremely cold winter they must 
have heating food and, like the Eskimo, the Alongol requires meat and much 


The entire life of the Mongols is the product of their environment, which 
is and has been dependent upon climate and geography. The coiintry is too 
dry, cold and windy for agriculttire, except in the southern grasslands. Con- 
ditions forced them to become a pastoral people, and nomadic. They depend 
entirely upon their flocks and herds for the necessities of existence. Their 
independence, love of sport, hospitality and admiration for the strenuous ele- 
ments of hfe are a direct resxolt of the conditions under which they live. 

In the real desert, sheep, goats and camels are the only animals that can 


exist in numbers upon the sparse, dry vegetation. In the grasslands, ponies 
and cattle are to be seen in considerable herds. The sheep and goats are 
gathered at the yurt at about ten o'clock in the morning for milking; then they 
are driven to the grazing grounds where they remain until dark, when they 
are again brought back for the night. Women and children do most of the 
herding. I have seen kiddies not more than five or six years old tending a 
flock of sheep in a most efficient way. 

A Mongol's wealth is always estimated by the number of animals which 
he possesses. Even the poorest natives own at least one pony and a few sheep 
or goats. During all my years in Mongolia, I have seen only one beggar on 
the plains, and he rode a very decent pony ! Before the Russian Soviet Govern- 
ment assiimed control of Outer Mongolia, the princes had herds of ponies, 
camels, sheep and goats, nvimbering many thousands. Cattle are compara- 
tively few. 

Although Mongols are necessarily nomads, they move only within certain 
prescribed limits. At a meeting of the inhabitants of a district with their 
officials in the early spring, grazing grounds are allotted to each family or 
village. They must adhere rigidly to these allotments. How many times they 
move depends entirely upon the grass. If mountains are near-by the Mongols 
go well up into the valleys during the stunmer, where there is better grass and 
water. In the winter they return to the plains. There are many parts of the 
desert where grazing is fairly good but where water is not to be had. Usually 
the natives go there in the winter, depending upon snow for water. They can- 
not remain in the mountain valleys dining the winter, for the snow lies so 
deep that their stock would be unable to dig down to the grass beneath it. On 
the plains the high winds sweep large areas bare, piling the snow into drifts. 

Since animal droppings (argul) are the only fuel and these bum rapidly, 
a considerable quantity is accumulated for winter use. Sheep dung is made 
into large bricks during the summer, and these are frequently piled about the 
yurt as a wall. This acts as a windbreak for the dwelling, a corral for the ani- 
mals and a fuel supply. If a family moves away from such a place in the 
spring, leaving a quantity of unused argul, other Mongols do not take it, as 
they know the owners will return. 


A Mongol's real home is the back of a pony. He is uncomfortable on the 
ground. His great boots are not adapted for walking and he is so seldom on 
foot that to walk a mile is piinishment. To go only a hundred yards or so he 
will jtunp on his pony, which always stands hobbled within reach. Children 
learn to ride in infancy. Each year, in the spring, a juvenile race was formerly 
held at Urga. Boys and girls from four to six years old were tied on ponies 


and rode at full speed over a mile-long course. If a child fell off, it received but 
little s\Tnpathy, and was strapped on again more tightly than before. A 
Mongol has no respect whatever for a man or woman who cannot ride, and 
nothing will win his admiration so quickly as good horsemanship. Mongols 
usually ride either at a trot or a full gallop. They use a broken snaffle bit and 
ride with a loose rein, always swinging the whip, which is a short stick with a 
lash at the end. 


Ponies are fairly cheap in Mongolia, but not extraordinarily so. Racing 
is almost a business and if a native owns a fast pony he is a lucky man. He 
goes to the annual field meets at all the temples in his neighborhood and will 
race for a sheep or a goat in the interim, or just for sport. The races are really 
endurance contests. Five to ten miles is the usual distance and I have known 
some races to cover twenty miles. The ponies are ridden by boys twelve or 
fourteen years old, who beat their moimts from start to finish. 

Next to horsemanship, the ability to shoot is most admired by a Mongol. 
Almost every native possesses a flintlock gun with an enormously long barrel. 
Its effective distance is hardly a hundred yards and they seldom shoot even at 
that range. They do not shoot offhand. Mr. Larsen speaks of the ability of a 
Mongol to shoot from a galloping horse, but I never have seen one even attempt 
it. Two long sticks are attached to the barrel on either side and these are used 
as a rest. WTien carrying the gim the sticks are folded back on either side of 
the stock. The Mongols never ceased to talk about the ability of our men to 
shoot running antelope offhand at three or four hundred yards. 


Every year the natives of each district in Mongolia hold a field meet. 
Usually these are at temples; sometimes at obos. Horse-racing, wrestling, 
shooting and rarely archery contests are the order of the day. Mongols 
gather from long distances, dressed in their finest clothes, and sometimes 
remain for a week or more. 

In 1919 I described a field meet in the forests north of Urga. The descrip- 
tion may well be quoted here as a typical example of a small meet : 

"The Terelche Valley meet was held on a flat strip of ground just below 
our camp. As my wife and I rode out of the forest, a dozen Mongols swept by, 
gorgeous in flaming red and streaming peacock plumes. They waved a chal- 
lenge to us and we joined them in a wild race to a flag in the center of the field. 
On the side of the hill sat a row of lamas in dazzling j^ellow gowns; opposite 
them were the judges, among whom I recognized Tserin Dorchy, though he 
was so bedecked, behatted and beribboned that I could hardly realize that it 



B. VLkl AMI LilKKAL H pR LA.\1B>. Klll^ AMi LAHh^. 





was the same old fellow with whom we had lived in camp. (I presume if he saw 
me in the clothes of civilization he would be equally surprised.) 

"In front of the judges, who represented the most respected laity of the 
commtinity, were bowls of cheese cut into tiny cubes. The spectators con- 
sisted of two groups of women who sat some distance apart in compact masses, 
the 'horns' of their headdresses almost interlocked. Their costumes were 
marvels of brilliance. They looked like a flock of gorgeous butterflies which 
had alighted for a moment on the grass. 

"The first race consisted of about a dozen ponies, ridden by fovirteen-year- 
old boys and girls. They swept up the valley from the starting point in full run, 
hair streaming and uttering wild yells. The winner was led by two old Mongols 
to the row of lamas, before whom he prostrated himself twice and received a 
handful of cheese. This he scattered broadcast as he was conducted ceremo- 
niously to the judges, from whom he returned with palms brimming with bits 
of cheese. 

"Finally, aU the contestants in the races, and half a dozen of the Mongols 
on horseback, lined up in front of the priests, each one singing a barbaric chant. 
Then they circled about the lamas, beating their horses imtil they were in full 

' 'After the race came wrestling matches. The contestants sparred for holds 
and when finally clinched, each with a grip on the other's waistband, endeav- 
ored to obtain a fall by suddenly heaving. When the last wrestling match 
was finished, a tall Mongol raised the yellow banner and, followed by every 
man and boy on horseback, circled about the seated lamas. Faster and faster 
they rode, yelling like demons, and then strung off across the valley to the 
nearest yurt. 

"This love of sport is one of the most attractive characteristics of the Mon- 
gols. It is common ground on which a foreigner immediately has a point of 
contact. The Chinese, on the contrary, despise all forms of physical exercise. 
They consider it 'bad form' and cannot understand sport which calls for violent 
physical exertion. They prefer quiet walks carrying their pet birds in cages 
for an airing; to play a game of cards; or, if they must travel, to loll back in a 
sedan chair with the curtains drawn and every breath of air excluded."^ 


The Mongols are a strong race. Their country itself is exceedingly healthy, 
and with ordinary attention to hygiene there woidd be little illness, but they 
are so ignorant and imcleanly that certain diseases are common. I have dis- 
cussed this subject more ftdly in Chapter XXXII. 

'Andrews, Roy Chapman. 192 1. "Across Mongolian Plains." D. Appleton & Co., New York. 
Pp. 158-159- 


One seldom sees sick Mongols. I suppose one of the reasons is that if a 
person is very ill the relatives simply decamp and leave the invalid to die. 
Believing that evil spirits take possession of a body as soon as life is extinct, 
they are extremely loath to have anyone die in their yurt. I have often seen the 
mute evidences of a desert tragedy — a skeleton lying beside the dead ashes of a 
fire ; near-by a wooden bowl with a little food ; there, the circular mark left by 
the yurt. The stor>^ was plainly told. The person was about to die and the 
other members of the family had moved to new grazing groimds, leaving the 
invalid to pass th.e last moment of his life alone in the desert. 

The Mongols are very superstitious about human remains. Under no 
circumstances will they touch or distvub a skull or skeleton. As soon as a per- 
son dies the body is dragged off to a considerable distance and left to be 
devoured by the dogs, wolves and birds. Sometimes the corpse is placed upon a 
cart which is drawii rapidly over rough ground. At some point the body 
falls off. The driver does not look back for fear that he will be followed by the 
evil spirits of the dead. At Urga there are several places not far from the city 
where corpses are left to be devoured by the dogs, and I managed to obtain a 
fine series of skulls for anthropological study. 

Venereal disease is difficult to check because of the promiscuous habits of 
the Mongols. They are unmoral rather than immoral ; they live like untaught 
children of nature, and modesty, as we conceive it, does not enter into their 
scheme of life. Adultery is openly practiced, apparently without prejudice to 
either party, and chastity is not a virtue. Traveling lamas very frequently 
demand a woman when they stop for a night in a yurt and they are seldom 

Life in the desert and on the plains of Mongolia is much like that of our 
own west in the pioneer days. Similar environment has developed similar 
customs. As in western America, hospitality is a law in Mongolia; assist- 
ance to a traveler is taken as a matter of course. 


When one comes to a Mongol yurt, he enters, sits dowTi beside the fire 
feeling sure of his welcome and is helped from the common pot. He may stay 
a day or several days without thought of payment. Every Mongol knows 
that he himself will ask for hospitality many times dxiring the year and thus he 
is ready to offer what he has to other travelers. 

I have often had Mongols ride several miles to bring my ponies to camp or 
tell me in which direction they had strayed ; they would expect as much them- 
selves in similar circumstances. To be left without a pony is a serious matter, 
for the distance between wells in the desert is often great. Horse stealing is a 
capital crime. In the days before the Soviet dominance, if a Mongol reported 


that a pony had been stolen, soldiers took up the trail and followed it until 
they ran the thief to earth ; usually he was shot at once. 


Next to ponies, dogs are probably the Mongol's most valued possession. 
The large Tibetan mastiff is the usual breed but smaller mongrel dogs are 
found ever3rwhere. All are exceedingly savage. They make excellent watch- 
dogs for yurt or caravan and are trained to attack on sight. It is dangerous to 
approach a yurt unarmed. In Urga the dogs eat human remains and are fed 
by the lamas. It is a crime to kill a dog. I have never seen a dog inside a yurt. 
The owners do not pet them, for savageness is a virtue. 

I have had several very narrow escapes from being killed and eaten by 
Mongol dogs, for when they are hungry they will attack a man like wolves. If 
one dog of a pack is injured and yelps, the others will tear it into pieces 
instantly. Doctor Berkey was attacked and probably would have been killed 
had he not shot the dog; almost all the men of the Expedition had more or 
less narrow escapes. 


Every newcomer in Mongolia is impressed by the rapidity with which 
news travels great distances ; it is often believed to be due to some telepathic 
method of communication which has been developed by the natives. It is 
true that news does travel in an amazing way, but I believe that the explana- 
tion is quite simple. 

The wells all over MongoUa are the natural meeting places and concen- 
tration points. Here the Mongols gather to water their stock and to gossip. 
If a traveler is near a well he wiU always ride over to see who is there and to 
hear the news. There is little to talk about and the slightest novelty is dis- 
cussed and rediscussed for hours. Very often a Mongol wiU ride forty or fifty 
miles to carry news to some of his friends; these in turn send it on to other 
ynrts. A fifty-mile ride is nothing to a Mongol. He knows that he always wiU 
find a welcome at any ywrt. He seldom has business of such importance that 
it cannot wait a few days while he disperses a choice bit of gossip. 


The Mongols have a direction sense which is most amazing. I often have 
been hunting gazelle on plains where there seemed nothing to serve as a land- 
mark. I might drop an animal and leave it for an hour or so. With a quick 
glance around my Mongol would fix the spot in his mind and dash off on a chase 
which might carry us back and forth toward every point of the compass. 
When it was time to return he would take us back unerringly to that single 


spot on the open plain. Of course, I learned to note the position of the sun, 
the character of the ground and the direction of the wind, but only by years of 
training can a white man hope even to approximate the Mongols' skill. They 
have been bom and reared upon the plains and have the inheritance of un- 
known generations whose life depended upon their ability to come and go 
at will. To them the hills, the sun, the grass, the sand — all have become the 
street signs of the desert. 


This wild free life of the plains has made the Mongol exceedingly inde- 
pendent. He relies entirely upon himself, for he has learned that in the strug- 
gle for existence it is he himself that covmts. Of the Chinese the opposite is 
true. His life is one of the community and he depends upon his family and his 
village. He is gregarious above all else and he hates to live alone. In this 
dependence upon his fellow men he knows that money counts — and there is 
very little that a Chinese will not do for money. We fovmd that the personal 
equation enters very largely into any dealings with a Mongol. If he likes you, 
remuneration is an incident. If you do not appeal to him, money tempts him 
but little. 

Although the Mongols are lazy under ordinary circumstances, they are 
not always so. Laziness is largely a product of their pastoral pursuits. Herd- 
ing sheep and goats requires little exertion, but at certain times their life 
demands extreme exertion and then the energy and endurance which they dis- 
play are amazing. I believe that the Mongol of to-day is quite equal to the 
warrior of Genghis Khan's time, as far as hardihood and endurance are con- 
cerned. Moreover, the present-day Mongol is an excellent fighter, as recent 
events have shown. 

In no part of the country which I have visited did the Mongols make any 
attempt at agriculture, but I have been told that a little farming is done in 
certain places in western and southern Mongolia. They sell a good deal of 
wool to Chinese traders, and imtil 1926 the International Export Company 
bought many thousands of sheep to be driven alive to Kweihwating or Hailar, 
where they were killed and packed for shipment abroad. 


The Mongols barter with the Chinese for those simple necessities of life 
which they themselves do not produce. Sheep and goats give them almost 
all they need. In the winter they dress almost entirely in sheepskins, but 
their summer garments are of cloth. From the Chinese they buy cloth, boots, 
tea and tobacco. The only things that I have seen them manufactiu-e are felt 
and silverware. 


I -«fe.<<i 



; — - '^"I^^^ 


The goats are held in Hne by a long rope which is looped around the neck of each animal. 




Felt-making is a rather interesting process and Shackelford photographed 
operations at Tsagan Nor. The sheep to be sheared were tied nose to nose on 
a long rope. Then one by one they were thrown and the wool roughly cut off 
with huge iron shears. Meanwhile two men were twisting the wool in loose 
ropes. Then it was taken to a flat plain above the yurts where a large strip of 
cloth had been spread upon the ground. Two old women spread a thick layer 
of wool upon the cloth and thoroughly soaked it with whey left from cheese- 
making, afterward spreading another strip upon it. This "wool sandwich" 
was rolled on a long pole, wrapped in a thin cloth and tightly bound. Ropes 
were fastened to the projecting ends of the pole, and a Mongol mounted on a 
camel dragged the cylinder behind him over a smooth path for several hours. 
Then it was again wet with whey and rolled. When the cylinder was finally 
opened the wool had been firmly pressed into a thick strip of felt. All that 
remained was to dry it in the sun and sew the edges with twine to prevent 


The Mongol women have a great deal more independence than those of 
China, but share in the work of herding the stock as well as attending to the 
domestic duties of cooking, making cheese and butter, and caring for the yurt. 
The children, as I have remarked, are put to work almost as soon as they can 
walk and pursue their business of herding goats and sheep in a most serious 
manner. The Mongols appear to have a good deal of affection for the children 
and take much pride in their ability to ride, wrestle or shoot. 


The men universally use snuff. When a Mongol comes into camp or sits 
down, beside the fire in a yurt, his first proceeding is to pass his snuff bottle to 
each person. It is exceedingly impolite to refuse to accept it, but it is not 
necessary to actually take the snuff. One can take the bottle in both hands, 
put it to one's nose and pass it on to one's neighbor. If the visit is an official 
one, or a favor is about to be asked, the visitor usually produces a light blue 
strip of sUk, called a hata, which is spread out and presented on outspread 

The abrupt way in which Mongols enter a tent or yurt is rather discon- 
certing at first. There is no preliminary knocking or asking permission; they 
simply come in and sit down. They invariably leave with the owner if he goes 
out. In Urga it was always difficiilt for foreign women to accustom them- 
selves to this habit. A Mongol will enter a house as tinceremoniously as 
though it were a yiort on the plains, and make his way about even into the bed- 
room without a word of warning. 


When a gift is presented to a Mongol, even if he is greatly pleased with it, 
he simply accepts it stoically without a sign of appreciation. I have some- 
times seen one put up a thumb, as do the natives all over the Orient, and say 
sat (good), but this is only when they are exceedingly pleased. Still they are 
not imappreciative and vindoubtedly do have a sense of gratitude. It is merely 
a custom of the coimtry. 


Since the Mongols are essentially nomadic they have developed no arts 
or manufactories. They can stay in one spot only as long as the grazing lasts. 
They must not hamper themselves with unnecessary household goods, or it 
would be too difficult to move. Even though, under Genghis and Kublai 
Khan, they conquered half the then-known world, they left nothing construc- 
tive behind them. They had nothing to give. At that time theirs was the 
same pastoral existence that it is to-day. Culture, art, architecture, did not 
enter into their lives. These could not be developed in a nomadic people liv- 
ing a wild, restless life upon the plains and deserts, with the struggle for exist- 
ence against the forces of nature ever present in their minds. 

If a Mongol of Genghis Khan's time should suddenly drop into the middle 
of Mongolia to-day, he would be perfectly at home. He would find that the 
ever>'day business of life, except in a few minor particulars, is carried on 
almost the same as in his day seven hundred years ago. There has been no 
reason to change the fundamentals of existence. The environment is very much 
the same. It is still a pastoral life. 


The only thing that has altered radically in the Mongol race is the spirit 
of the people and their religion. A Mongol of Genghis Khan's time would find 
them no longer a race of warriors. He would find that two thirds, at least, of the 
male population had donned the yellow and red robes of lamas ; that they had 
become dissolute hvunan parasites. It would be difficult for him to adjust 
his mental perspective to such a state. It is totally incongruous to a people 
who live upon the plains and deserts combating the forces of nature for their 
very existence. 

There were several contributing causes to the decay of the Mongol race, 
but the primal factor was the introduction of lamaism about 1290. Before 
this they had been Shamanists, worshiping the spirits of nature that lived in 
the rocks and trees and moimtains. Lamaism became the reUgion of the state. 
Its teachings are against war, learning, enterprise, ambition. Fostered by the 
Chinese, who realized its value in subjugating a warlike race, it obtained such 
a powerful hold over the superstitious nomads that it became the paramount 


factor in their lives. The authority of chieftains and generals gave way to 
that of Buddhist priests. 

Custom decreed that the first-bom son of every family should become a 
lama. Sometimes all the male children joined the priesthood. Many of them 
remained permanently in the temples which had sprung up over all the coun- 
try. Those who returned to their yiu-ts had learned habits of slothfulness and 
were periodically at the call of the ruling lamas of their district. Were one to 
enter ftdly into a discussion of the evil effects of this religion, he could trace 
to that soiirce virtually all the major factors in the disintegration of this virile, 
once-glorious race. Under Soviet dominion lamaism is frowned upon. The 
Russians have not allowed the appointment of another Hutukhtu to succeed 
the one who died in 1923. They have curtailed the power of the lamas and 
are doing their utmost to destroy the hold of the religion over the people. 
They can make but little progress except by force, I feel sure. 


It is not the purpose of this chapter to discuss the past history of the 
Mongols or their political condition. Still, the 1922 status of the coimtry and 
the changes of the last few years are so little known to the western world that a 
brief sketch of the recent political events may be of interest. Most of them 
happened while I was in Mongolia. 

Until the overthrow of the Manchu regime in China in 191 1, and the 
establishment of the present republic, there were no particularly significant 
events in recent Mongolian history. At that time the Russians, wishing to 
create a buffer state between themselves and China, and especially to obtain 
commercial privileges in Mongolia, aided the Mongols in rebellion, ftimishing 
them arms and ammiuiition, and officers to train their men. 

A somewhat tentative proclamation of independence for Outer Mongolia 
was issued in December, 19 11, by the Huttikhtu and nobles of Urga, and the 
Chinese were driven out of the country with little difficulty. Beset with internal 
troubles, the Chinese paid but scant attention to Mongolian affairs vmtil news 
was received in Peking, in October, 191 2, that M. Korostovetz, formerly Rus- 
sian Minister to China, had arrived secretly in Urga and, on November 3, 
191 2, had recognized the independence of Mongolia on behalf of his Govern- 

It then became incumbent upon China to take official note of the situa- 
tion, especially as foreign complications could not be faced in view of her domes- 
tic embarrassments. Consequently, on November 5, 1913, there was concluded 
a Russo-Chinese agreement wherein Russia recognized that Inner Mongolia 
was tmder the suzerainty of China, and China on her part admitted the auton- 
omy of Outer Mongolia. The essential element in the situation was the fact 


that Russia stood behind the Mongols with money and arms, and China's hand 
was forced at a time when she was powerless to resist. 

Quite naturally, Mongolia's political status has been a sore point with 
China and it is hardly surprising that she should have awaited an opportiuiity 
to reclaim what she considered to be her own. This opportunity arrived with 
the collapse of Russia and the spread of Bolshevism, for the Mongols were 
dependent upon Russia for material assistance in anything resembling military 
operations, although as early as 1914 they had begvm to realize that they were 
ctiltivating a dangerous friend. The Mongolian army, at the most, numbered 
only three or four thousand poorly equipped and undisciplined troops who 
would require money and organization before they could become an effective 
fighting force. 

The Chinese were not slow to appreciate these conditions, and General 
Hsu Shu-tseng, popularly known as "Little Hsu," by a clever bit of oriental 
intrigue sent four thousand soldiers to Urga under the pretext of protect- 
ing the Mongols from a so-called threatened invasion of Buriats and brig- 
ands. A little later he himself arrived in a motor car and when the stage was 
set brought such pressure to bear upon the Hutukhtu and his cabinet that 
they had no recourse except to cancel Outer ]VIongolia's autonomy and ask 
to return to their former place under Chinese rule. This they did on Novem- 
ber 17, 1919, in a formal Memorial addressed to the President of the Chinese 
Republic. Naturally the President graciously consented to allow the prodigal 
to return and "killed the fatted calf" by conferring high honors and titles upon 
the Hutukhtu. Moreover, he appointed the Living Buddha's good friend (?), 
Little Hsu, to convey them to him. 

Such happy conditions did not last long. The Chinese soldiers in Urga 
and other parts of Mongolia behaved so outrageously in every way that the 
Mongols could not endure their t>Tanny. They entered into communication 
with Baron Ungern-Stemberg, who agreed to drive the Chinese out of IVIongolia 
if they would give him three million dollars and allow him to use Urga as a 
base to prepare for his campaign against the Bolsheviks at Kiakhta. He 
attacked Urga on January' 30, 192 1 , with a force of about five thousand. These 
included four hundred Russian Cossacks and about a thousand Tartars, the 
rest of the army being made up of Buriats, Tibetans, Mongols and Japanese. 
Urga was captxired with little difficulty by Ungem, who lost only sixty killed 
and one hundred and fifty wounded. Then, and subsequently, eight or nine 
thousand Chinese soldiers were killed. Baron Ungem sent parties to Ulias- 
sutai, Kobdo and other points where there were known to be Chinese soldiers, 
Jews or Bolsheviks. Their massacre was wholesale. 

Eventually Ungem left Urga to attack the Bolshe\'iks at Kiakhta, but was 
defeated and himself killed. The Reds then came into Urga with the assurance 


that they were there to protect the Mongols. By the winter of 1922 they had 
proclaimed a Mongolian Republic and changed the name of Urga to Ulan 
Bator, the "Red City." A Mongol Cabinet was appointed in which each Min- 
ister was complemented with a Russian adviser. The Mongols were virtually 

Russian influence has been gradually extended by the introduction of 
Buriats (Mongols bom in Siberia and educated as Russians) into all official 
positions. At the present time (1930) Outer Mongolia is in all practical ways a 
Russian province. 

The establishment of the Bolshevik regime is, I believe, the last act in 
the tragedy of Mongolia. The doom of the Mongols, as a race, is sealed. 



The Tsagan Nor region offered a most fruitful field of investigation for 
every branch of the Expedition. Shackelford and I were busy with zoological 
studies and photography at the lake, while the geologists worked night and 
day upon their map of the section from Uskuk to Baga Bogdo. Granger reaped 
a rich and varied harvest of small mammal fossils. 

During a single day's prospecting in the Hsanda Gol formation at the 
"Grand Canyon," Granger picked up one hundred and seventy-five jaws and 
skulls of carnivores, rodents and insectivores. He maintained his lone camp 
at Loh and from there made an excursion to Uskuk, where he discovered the 
fine skeleton of a small dinosaur in the Cretaceous Ondai Sair formation. It 
was so interesting and complete that I sent Shackelford to Loh to photograph 
the uncovering and removal of the specimen. 

The skeleton proved to represent a new family of iguanodonts to which 
the name Psittacosauridae has been applied by Professor Osbom. The Uskuk 
specimen has been named Protigiianodoji mongolietise by Professor Osbom. 
It is a short-skuUed, parrot-like beaked, bipedal, cursorial dinosaur, closely 
allied to another type, Psittacosaurus mongoliensis, fovmd a few weeks later in 
the Oshih basin. 

On July 30, Colgate and I went to Loh to visit Granger and Shackelford. 
We foimd Shackelford sufTering from an attack of hives, and he was very much 
disgusted when I assured him that he did not have some really interesting 
disease such as "Gobi fever." This was the only case of illness on the Expe- 
dition during the entire summer. 


Two days earlier Shackelford made a most important find. He was walk- 
ing down a dry stream-bed near camp and actually stumbled over the distal 
end of a humerus of a Baluchitherium. The huge bone had weathered out of 



the perpendicular bank and dropped to the stream-bed. Then we realized that 
the Mongol reports of "bones as large as a man's body" had some basis in fact. 
Our first evidence that Baliickitherium had inhabited this region during the 
Oligocene was obtained when Doctor Berkey found a calcaneum and frag- 
ments of other tarsal or carpal bones near Iren Dabasu. 

I went with Granger to examine the spot and we searched the stream-bed 
and the surroimding country carefully, but not another trace of the skeleton 
could be found. Later we went to one of the near-by knolls to prospect for 
other fossils. The groimd was thickly strewn with teeth, jaws, and fragments 
of skulls, mostly of rodents, but the preservation was bad. Remains of two 
new genera of Bathyergidas were prominent because of their size and abund- 
ance. Dr. W. D. Matthew described these as Tsaganomys altaicus and Cyclo- 
mylus lohensis. He remarks: "This Tsaganomys appears, if properly referable 
to the family, to be the first fossil record of Bathyergidae, hitherto known from 
the recent Ethiopian fauna. It is by no means close to the living genera and 
should perhaps be distinguished as a separate subfamily, Tsaganomyinae, on 
the short massive proportions of sktill with heavy forward pitch of occiput, 
wide differences in otic region and some rather minor differences in teeth. It 
suggests the Asiatic ancestry of the family, although it cannot be considered 
as even approximately ancestral to the living genera."' 

Although many of these skulls were exposed, only a few were worth 
removing. The bones were so broken and decayed that they could be blown 
away like dust, and no one less expert than Granger could have taken them 
out at all. 


Colgate and I spent only a day at Granger's camp, and on the way back 
to Tsagan Nor visited a line of traps which I had set out for a large diurnal sand 
rat, RJiombomys, in a dry river-bed near the "Grand Canyon." At this point 
the plain was low and thickly covered with a growth of so-called tamarisk, 
Salicornia herbacea. Some of the trees reached a height of fourteen feet but the 
majority were not more than five or six feet high. The stream-bed led straight 
down from the range of low lava-capped mountains at the northern margin 
of the basin and must have been largely responsible for the deeply eroded 
"Grand Canyon." It indicates, with other evidence, that at some time in the 
not far distant past there was a period of much greater moisture in this region. 
The "tamarisk" are most abundant where the waters of the stream debouch 
upon the plain. Very probably these trees obtained such a firm foothold dur- 

' Matthew, W. D., and Granger, Walter. 1923. "New Bathyergidas from the Oligocene of MongoUa," 
Amer. Mus. Novitates, No. loi, Dec. 28, p. 4; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. 
Nat. Hist., I, No. 19, pp. 1-5, 1926. 


ing the moist period that they can still send their roots deep enough to find 
sufficient water to keep them aHve. It is significant that no young growth 

East and northeast of the lake there is a low area of great playa-tiofts, or 
specifically "sand hillocks." These are formed where shrubs and low bushes 
net the soil with their roots and form a small hummock at each plant. Drifting 
sand is entrapped by the growing vegetation and the hillock is, in part, thus 
built up. Some were six or eight feet high and usually conical in shape. Most 
of them were covered by a thorny shrub, Nitraria schoberi, which we called 
"dime berry" because of a delicious red fruit which it bears. This was the only 
edible fruit we found in the Gobi, but it was so abundant in most places that 
in August, when it ripened, we were usually well supplied. The only disad- 
vantage was the inroads cooking the fruit made upon ovu" sugar. 


The geologists in the progress of their map circled Tsagan Nor and crossed 
the belt of live sand-dunes which lies between the lake and Baga Bogdo. Sub- 
sequently we all went over the sand-dimes on camels to investigate a fossil 
exposure which they discovered. Sand-dunes are an accompanying feature of 
almost all the lakes which we saw in the Gobi. Professors Berkey and Morris, 
1927, write as follows regarding their formation: 

"The wind-bome dust and sand forms deposits of several types, depending 
upon the abundance and quality of the sand, the direction, strength and con- 
stancy of the wind, the topography of the region, the rainfall and the vege- 

"Where the supply of sand is abundant, true dunes arise on the lee-side 
of the source, and march in the direction of the wind. At the foot of Baga 
Bogdo, a long scimiter-shaped area of big dunes curves eastward for more than 
thirty miles from the dissected sediments and dry stream-beds, which furnish 
the sand. The shape of the dunes varies with the wind. Most of the time, the 
westerly wind is so strong and constant that the dimes approach the shape of 
typical barchans. But during violent storms from the east, the shapes become 
sinuous and irregular, though the crestlines still lie rudely north and south, 
transverse to the direction of the prevailing winds. 

"It was our good fortune to cross the dune belt six times in 1922, twice 
during high winds which permitted observation of the movement of the sand. 
A thin sheet of sand-fiUed air leaped from the crest of the dune, like spray 
from the crest of a big sea wave. The sand spray dropped upon the leeward 
side of the dune, down which it ran in long rills that looked like syrup spilling 
down the face of the dune. New riUs started from the crest, or near it, every 
few seconds. We looked for eddies on the leeward side, and saw some, but 


RESTORATIO?J OF Baluchithcrium under the direction of henry FAIRFIELD OSBORN. 
Painted by Charles R. Knight, 1923. 


■* i -'■] 


B. SKULL OF Baluchitlwriiiin discoverf.d in 1922. 


they were small, and were not important factors in shaping the advancing 
front of the dune. The chief factor was the simple delivery of sand from the 
back slope to the front. 

"With less abundant supply the sand forms thin sheets over the level 
surfaces, especially on the upland above the valley, or hollow, from which the 
sand comes. It climbs the windward slopes of hills, blanketing them against 
effective wind erosion."^ 

When I visited the sand-dunes, I was amazed to find that in the bottoms 
of the valleys between the dunes there was a considerable quantity of very 
long, coarse grass, and tall bush sweet-peas bearing beautiful piuple flowers. 
It was evident that water could not be very far below the surface, and in sev- 
eral places Shackelford scooped out a "well" with his hands, finding drinkable 
water within fovu- or five feet. This is probably due to an accumulation of rain 
which has drained to these low spots between the dunes. Baga Bogdo is high 
enough to break up the clouds and precipitate moisture, and very often we 
watched heavy rain falling on the mountain slope which never reached us ten 
miles away on the north shore of the lake. 


Except for a few light showers we had no rain from June 25 until August 
3, the night Berkey and Morris returned to camp after completing their map. 
They arrived just at dark, very tired but satisfied. They had mapped, both 
topographically and geologically, an area of eight hundred square miles in a 
strip extending from the northern limit of the basin at Uskuk to the southern 
margin at Baga Bogdo. 

Berkey and Morris were full of interesting details of their experiences on 
the opposite side of the lake. They had been amazed at the remarkable size 
of the alluvial fans which debouch from the mouth of every canyon. They 
ascended one of the fans and found that it was ten miles from base to crest and 
two thousand feet high. They had encoimtered sandstorms which almost 
smothered them and had discovered somber canyons with beautiful cascading 
streams. It was like listening to a true fairy story, and the camp did not go 
to rest until the early hours of the morning. 

Most interesting was the report of Berkey and Morris of a new fossil- 
bearing exposure at the foot of Baga Bogdo, and a fragmentary collection of 
bones. Among the bones I recognized a large cervid and a mastodon; this 
was sufficient to identify the deposit as Pliocene or Pleistocene and to indicate 
that we must give it fxorther examination. For the formation Berkey and 
Morris had suggested the name Hung Kureh. 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, II, pp. v-xxxi, 1-475. 



The next evening, while we were silent with awe witnessing the most 
glorious sunset that I have ever seen, a black car slipped into camp from the 
north, carrying Granger and Shackelford. Their work at Loh was finished 
and they came down with all their gear and specimens. Their faces were 
radiant with suppressed news and when the celestial display was ended and 
the soft evening light had enshrouded the rugged outUnes of Baga Bogdo in a 
robe of delicate purple, they produced a prize exhibit — newly discovered bones 
oi a. giQXit Bahichitherium! These had been found after Granger and Shackel- 
ford had broken camp and were on their way to the lake. So it happened all 
through the summer! We had only to pack and make ready to leave a fossil 
bed to discover some priceless specimen ! 

Granger and Shackelford had decided to walk through a still uninspected 
pocket in the badlands and to have Wang, their Chinese chauffeur, drive the 
car ahead to a promontory two miles to the south. After a little, Wang decided 
to do some prospecting on his own accotmt. Almost immediately he discov- 
ered a huge bone in the bottom of a gully that emptied into a larger ravine. 
It was the end of a humerus of a Baluchitherium, and other parts were visible, 
partially embedded in the red earth. 

We had impressed upon all our natives the importance of leaving a bone 
in situ, and Wang waited until Granger arrived. Then they found one side of 
a lower jaw and other fragments of the skeleton. The bones were so hard 
and free from matrix that they could be easily removed, and Granger brought 
everything to camp at once. 

We talked far into the night, and I awoke next morning filled with a 
desire to find more of the Baluchitherium. Granger could not leave camp, as 
he was packing fossils to go with the camels which we wished to start east- 
ward as soon as possible. Therefore, Shackelford, Wang and I drove back to 
Loh to give the gully a more careful inspection. Shackelford and Wang began 
to dig a trench where the jaw had been found while I prospected the sides of a 
small dividing ridge. Arriving at the crest of the ridge, I looked down the other 
side and instantly saw fragments of bone half covered with loose sand in the 
bottom of the wash. I knew that the bone fragments were portions of a 
Baluchitherium because of the color, and leaped down the slope with a shout. 
Shackelford and Wang came on the run and in a short time we had unearthed 
several large fragments; then our fingers struck the end of a huge block in 
which a loose tooth was embedded. Evidently we had discovered a skull. I 
knew that it was time to stop, and we returned to camp with otir spoils, for I 
was too excited to do further prospecting that day. We burst in upon camp as 
the men were having tea. The next day a party retiimed to Loh to begin the 
work of removing the huge bones. 


The skeleton evidently had been buried at a spot that later became the 
summit of a ridge between two gullies. As the sediments weathered away, part 
of the skeleton rolled down one side of the ridge ; that was what Wang found 
the first day. The rest had slipped down the other side into the main wash, 
where I discovered it. We followed the ravine down to the mouth, rescuing a 
few fragments that had been carried away by rain action, and Shackelford 
discovered several important sections of the sktoll at least three hundred yards 
out on the plain. 

We sifted every square inch of sand, and, at the end of the four days 
which Granger required to remove the skull, we felt certain that nothing had 
been overlooked. The huge block was strengthened with burlap and packed 
in two sections in camel boxes preparatory to its long journey to New York. 
Professor Osbom has said: "The packing of this skull, its transportation across 
the desert of Mongolia, its preservation from bandits and from the unpaid 
soldiery, its journey to Peking, thence to the nearest port and finally its safe 
carriage to the American Museum, where it arrived absolutely unimpaired on 
December 19th, 1922 — these are among the great events of palaeontologic 
history." The restoration of the skull was begun immediately by Mr. Otto 
Falkenbach under the direction of Professor Osbom, who made a detailed 
study of the animal. 

First knowledge of Baluchitherium was given by C. Forster Cooper, 191 1, 
1913 and 1923, of Cambridge University, England. In 191 1 he discovered, in 
the Bugti Hills of eastern Baluchistan, two fossil aberrant rhinoceroses — one a 
small type which he named Paraceratherium, and evidences of an animal of 
gigantic size which he designated Baluchitherium oshorni. He fotmd only parts 
of the cervical vertebrae and foot bones and, although from the first he suspected 
its relationship to the rhinoceroses, he was unable to determine its kinship 
with certainty. His conjecttire was substantiated by the discovery by a Rus- 
sian palaeontologist, A. Borissiak, 191 5 and 1916, near Turgai in Russian 
Turkestan, of remains of a gigantic beast which he named Indricotheritim. 
Borissiak found well-preserved grinding teeth and parts of the skeleton which 
indicated that his specimen and Cooper's Baluchitherium were closely related 
or perhaps identical, but their affinities to other rhinoceroses were uncertain. 

Professor Osbom's study, 1923, justified the estimate of its original dis- 
coverer, C. Forster Cooper, that it was probably the largest land mammal 
known. He concluded that the Baluchitherium reached a height of seventeen 
feet at the shoulders and was about twenty-four feet in body length ; and that 
it had a long neck, stilted limbs and shoulders well elevated above the hips as 
is usually the case in tree-browsers. He is convinced that it had prehensile 
lips adapted to feeding on the herbage of the higher branches of trees, like a 
giraffe. With the elevated body-form and massive neck, the head, enormous as 


it is, diminishes in relative size, although far exceeding that of any existing 
mammal in absolute size. The caniniform adaptation of the incisor teeth is 
remarkable. They are veritable tusks, wielded by a sktall of surpassing size and 
weight and a neck of gigantic proportions, terrible weapons of defense and 

The finding of remains of a specialized mammal of such enormous size in 
regions so widely separated as northern India, Russian Turkestan and central 
Mongolia is an indication that during the Oligocene period, when it lived, cli- 
matic and environmental conditions must have been very similar over all 
central Asia. That the Bahichitheriiim never reached Europe or America is 
not surprising. It was too highly specialized to make the long journey to 
America, and Grabau has shown that it was cut off from Europe by the Turgai 
Straits, which separated Russo-Scandia from Pal-Asia during Oligocene 


On August ID all of us, with three Mongols and a cook, mounted camels 
for a trip across the sand-dunes to the Pliocene exposures, which had been dis- 
covered by the geologists and named the Hung Kureh formation. The forma- 
tion lies southeast of Tsagan Nor and is exposed in a bold white escarpment, 
facing west and north. Near the base of the bluff, the lowest beds visible are 
yellow sands which contain fossils. Above them are fine white sands and 
light gray clays, forming the face of the bluff, which is about two hundred 
feet high. The clays and sands are abruptly succeeded by a rubble of coarse 
pebbles which are only slightly rotmded. 

Having arrived at the exposures, we spent the afternoon himting fossils 
with only indifferent success, and at sundown set off for a spring two miles 
away where we were to spend the night. We found the cook there and dinner 
ready. The water was delicious, for it bubbled from a clean gravel basin and 
spread out in a series of threadlike rivulets bordered by a soft carpet of emer- 
ald-green grass. 

The next day, tmtil noon, we prospected the beds near camp but failed to 
find fossils. Apparently they are a somewhat later deposit, and we were 
anxious to find bones to assist the geologists in dating them. After tiffin we 
went back to the typical Hung Kiu-eh exposures. 

A few hours' work produced a sizable collection of specimens, although 
most of them were in a fragmentary condition. Gazella, Camelus, Hipparion, 
Castor, Cervus, and a mastodont were easily identifiable; most interesting of 
all were bits of eggshell and toe bones of Struthiolithus, a giant ostrich. This 

• Grabau, A. W. 1927. "A Summary of the Cenozoic and Psychozoic Deposits with Special Reference 
to Asia," Bull. Ceol. Soc. of China, VI, No. 2, 3, pp. 183-187. 



s I 

2 B 


< « 


huge bird was known only from the eggs which had been found in the loess of 
north China. The eggs are at least a third larger than those of the existing 

Late in the afternoon I had the good fortune to discover the perfectly 
preserved antler of a large deer. As I was inspecting a low, rounded ridge of 
fine yellow sand and clay, I saw some fragments of bone, and following them 
up found a slight discoloration. From its center the tip of what appeared 
to be a large tooth was just visible. The matrix was easily worked away 
and I exposed enough to show that the specimen was a shed antler as large 
as those of the American wapiti, Cervus canadensis. Granger had only 
two hotu-s in which to remove the specimen, but by skilful work with rice- 
paper and gtun arable, he took it out in sections and carried it safely back to 

Another day's excursion to the "Grand Canyon" produced a himdred jaws 
and teeth and skiills of small mammals, including rodents, insectivores and 
carnivores. We found a great quantity of fragmentary Baluchitheriiim bones, 
but the pieces were so broken that none of them was of value. They must have 
come from several skeletons which had been exposed and entirely disin- 
tegrated by weathering. Doubtless this locality will produce a more or less 
complete Bahichitherium skeleton sooner or later. 


Sunday, August 13, was the day appointed for leaving our beautiful camp 
at Tsagan Nor. We could already feel an autvimn sharpness in the morning 
air, and water-fowl were dropping into the lake only to leave for the south in a 
day or two. The sand-grouse had gathered into enormous flocks which swept 
in a gale of brown wings above our tents, flying usually from west to east 
parallel with the mountains. We must travel a thousand miles through vir- 
tually unknown country about which little information could be gleaned from 
the few Mongols we had seen. We expected to move independently of the main 
caravan trails ; if the terrain were fairly good we would have time for further 
investigation, but if the region proved to be impassable we might be compelled 
to tiim back and seek a different way out. I felt some concern, for we had 
only enough gasoline to take us comfortably to Kalgan. 

We planned to travel along the Kweihwating-Kobdo trail, which passes 
just north of Tsagan Nor, skirts the eastern end of Baga Bogdo and con- 
tinues past Artsa Bogdo fifty miles farther to the east. The caravan left on 
the day of our departure, with instructions to meet us at Artsa Bogdo. The 
Mongols reported Artsa Bogdo to be a particularly good district for ibex and 
bighorn sheep, and it was probable that we wovild also find sedimentary basins 
worthy of palseontological investigation. 



Twenty miles from Tsagan Nor wc halted near two small ponds bordered 
by marshland. They were alive with ducks and geese and we saw an oppor- 
tvinity to relieve the monotony of our antelope-meat diet. In an hour we got 
twelve graylag geese, eight mallard ducks, one whooper swan, three jack-snipe 
and two beautiful painted snipe, Rostratula beiighalensis benghalensis. The 
swan was winged-tipped by a single shot and we caught it only after a lively 
chase. I grasped it by the neck but relinquished it just as quickly, for the bird 
was able to strike such a powerful blow with its wings that it was really danger- 
ous to one's legs and arms. After photographing the swan we released it upon 
the largest pond; doubtless it would soon be able to fly again. Having tried 
to eat adult swans and found them as tough as rubber, I had no desire to risk 
any of the Expedition's teeth. 

A short distance east of the ponds we reached a series of small streams 
which drain from the northern slopes of Baga Bogdo. They were only two or 
three feet wide, but their steep banks gave us much trouble. We had to bridge 
them with turf -sods, and all the men were exhausted by the time the last stream 
had been crossed. The water-table lies so close to the surface that the cars 
were in almost constant danger of miring. 

We saw more surface water in this region than elsewhere in the Gobi. 
Some of the streams actually flow for miles across the plain. About thirty 
miles east of Tsagan Nor there was a large valley leading toward the northeast 
away from the mountains. Regarding the valley. Professors Berkey and Morris, 
1927, remark: "It must have been made by stream erosion where now no 
competent stream exists. This is another of the numerous evidences of changed 
climatic conditions. With the present rainfall it would be impossible to carve 
a valley of such magnitude. Therefore, the presence of these forms, demanding 
different conditions for their sculpturing, lends strong argiunent for a different 
climate in former times. There must have been more water, and there must 
have been greater streams — for valleys were carved out which are now slowly 
filling or are kept cleaned out by the work of the wind, which helps to remove 
the new deposits."^ 


The basin, watered so abundantly from the Baga Bogdo streamlets, was a 
veritable oasis in the desert. There were no trees, of course, but a carpet of 
emerald-green grass which contrasted strongly w4th the brown gravel hills on 
either side. Hundreds of sheep, goats and ponies drifted along the valley 
bottom and a score of yurts dotted the slopes. When I turned my back to the 

> Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. v-xx.xi, 1-475. 


desert and looked over the green meadow to the rugged slopes of Baga Bogdo, 
it was difficult to realize that we had not suddenly been transplanted to a valley 
of the American Rocky Mountains. 

The Mongols were most friendly, and we recognized several who had com- 
peted in the field meet at Tsagan Nor. They greeted us like old friends and 
gained much "face" with their less sophisticated neighbors by "explaining" 
the workings of our motor cars. 

We were all tired by the time the bad going of the valley had been crossed, 
but our reward awaited us in the form of a slightly rolling gravel plain as 
hard and smooth as a tennis court. The centuries old trail showed as a faint line 
for miles ahead. The short, stiff bimch grass was greener than that of the desert 
to the west, and isolated volcanic outcrops showed black against the sky. 


At about eight o'clock we camped in the picturesque canyon of a tiny 
stream which had cut its way through a low ridge of volcanic hills. The water 
was charged with mineral matter, but the indefatigable Shackelford dug a well 
in the loose sand and gave us drinkable water in an hour. Our day's run had 
been only forty-five miles, due to the bad going and a late start from Tsagan Nor. 

The next morning dawned clear and sunlit with a stimulating sharpness 
in the air. We drove across an oldrock floor of pre-Cambrian conglomerates, 
limestones and other rocks on a beautifiil hard trail. At either side craggy 
remnants of the ancient rock stood up two or three hundred feet in height, and 
the trail passed through gaps and saddles between them. We had a fruitless 
chase after two wolves, but they outwitted us among the rocks. We drove 
merrily on to sweep around a great promontory and view the rounded mass of 
Artsa Bogdo rising from a gently lifted, green-clothed plain. As we waited for 
the other motor cars to arrive, five wild asses accompanied by a small herd of 
gazelle trotted into view. This was our "farthest east" record for wild ass at 
the time, but subsequently we found that the limit of their range was the 
Gurbun Saikhan. 


At a long yurt beside the trail, an old hunchback Mongol told us that there 
were several native encampments at the base of Artsa Bogdo. With the field- 
glasses we coiild see yurts and started across country toward the largest village. 
Two Mongols herding camels near the trail saw us coming, deserted their 
charges and dashed for the movmtain. They reminded us of wild asses for they 
wovild ride some distance at ftoll speed, and then stop to gaze back at us. They 
never had seen or heard of a motor car and curiosity almost overcame their 


Although the mountain appeared to be not more than three miles away, 
we discovered that it was ten miles and the rise from the trail was one thousand 
feet. Our motors were boiling under the heavy loads on the long grade before 
we reached the nearest village. Panic was the order of the day. Men and 
boys threw themselves on ponies and dashed for the mountain. Women and 
girls, of less account, perforce took refuge in the ytirts, for we came too rapidly 
to make escape on foot practicable. 

When the cars reached the village, our Mongols entered a few of the yurts 
to assure the inhabitants of our friendly intentions. Soon one or two ventured 
reluctantly outside, and in half an hour most of the women and old men 
were assembled about the cars. The able-bodied men who had been watching 
from the shelter of the hills began to straggle back when they saw that the 
expected slaughter had not materialized. It was an illuminating commentary 
on the deathly fear which the bloody events of the last two years had inspired 
in these peaceful desert dwellers. 

The tents were pitched on a grassy slope half a mile above the village and 
right at the base of the mountain. Before us unfolded a magnificent panorama 
of desert and "badlands" and about the camp there was a delightful atmos- 
phere of cleanliness, height and freedom. We looked forward to an interest- 
ing fortnight, for the Mongols assured us that the mountains swarmed with 
sheep and ibex, and the badlands looked decidedly possible for fossils. 



The Baga Bogdo range of the Altai Motintains is separated from the 
Artsa Bogdo unit by lower ground, and is slightly offset to the south from the 
direct east and west line, but in general view the two appear to be a single con- 
tinuous chain. The western mass of Artsa Bogdo is high and rugged and 
separated by a "saddle" of lower hills from the eastern end, where the moun- 
tains are high but consist of rounded grass-covered summits with comparatively 
little rock exposed. 

We were camped at the base of the saddle on the edge of a deep gorge, cut 
by a stream straight through the motantain. Two wells in the bottom of the 
river-bed gave us delicious water, and in a day we were on friendly terms with 
our new neighbors. There were three groups of fifteen yurts each, separated 
by perhaps half a mile. Sheep, goats, cattle and ponies enjoyed the abundant 
grass, and ten miles away we could see the brown patches of camel herds feed- 
ing on the dry thorny vegetation of the desert. We soon learned the reason 
for the sharp demarcation of grassland and desert, for often we were drenched 
with rain which fell on the mountain slopes but seldom reached the desert 
plain a few miles away. 


The Mongols of the villages soon realized that not only had we brought 
much of interest into their dtdl lives but that we were ready to pay well for the 
assistance which they could give. On the first afternoon I engaged a lama gmde 
to take me shooting the following morning. I rose before the first gray light of 
dawn and, moimted on a tiny twelve-hands-two pony, rode up the dry stream- 
bed in the bottom of the canyon for about three miles. The air was clear and 
cold; quail, Coturnix, buzzed continually out of the long grass at our feet; the 
familiar call of red-legged partridges, or chukars, Alectoris, sounded on every 
side ; conies whistled constantly and a fox barked at me from the shelter of a 
rock. After our long weeks in the desert it was delightful to have my leg across 

a pony and to breathe the mountain air again. 



We came upon a herd of thirty female and young ibex shortly after we 
had climbed out of the canyon to the upper slopes of the hills. I was surprised 
at the color of the animals — a rich nut-brown, like roe-deer in the winter pelage. 
I did not have a shot, but a few hours later knocked over an old female as she 
dodged among the rocks on the rim of the gorge. Her long beard, short horns 
and yellow eyes made her appear very goat-like; her dark brown coat so per- 
fectly matched the color of the rocks upon which she lay that she was well- 
nigh invisible. 

I saw fifty-seven ibex that day and two bighorn sheep, but all were 
females or yotmg. Subsequently we learned that during the summer and imtil 
the rutting season the male ibex live among the high, rough peaks of western 
Artsa Bogdo, and only an occasional straggler wanders into the lower hills where 
we were hunting. 


When I returned to camp just before sunset I fotmd that the geologists, 
with Granger and Colgate, had prospected some extensive exposures of red 
beds between the mountain and the trail, without discovering fossils; also, 
that they had investigated an interesting mesa south of the trail. It is capped 
with black basalt, and the white and red sediments showing on its scarred 
flanks make a spectacular color combination. The absence of fossils was very 
surprising, for although the age of the sediments is unknown, the aspect is like 
that of some of the Tertiary beds near Tsagan Nor. 

I did not visit the mesa, but Doctors Berkey and Morris, 1927, remark 
concerning it. 

"On some of the basalt blocks on the white-clifTed mesa, we noted a strik- 
ing display of prehistoric pictiires. One group of these, found oh a large, flat 
fragment, carries the following: two men, one with bow and arrow, several 
animals, one of which was probably intended to represent a horse, another a 
reindeer, another possibly a moose, while still another is perhaps a fox. None 
of these animals, except the fox and the horse, is to be seen within several 
hundred miles of this region at the present time, yet it must be that the 
artist was familiar with them. Either he belonged to a tribe which came 
from a region where these animals were native, or else the region itself was 
occupied in his time by such animals. It is entirely reasonable to believe 
that the latter is the explanation, and that the artist pictured animals living 
■ in his vicinity. This interpretation would mean that the people who made 
these drawings were not of the Mongol race, and that the climate was other 
than it is now."* 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. v-xxxi, 1-475. 




SS*y^£k.. --.„ 




'»n K 


~' '; 


^> ■ % ^ 






/ 4- 

r, ;*•;*< Z:^-,^... v. ^- 

B. MO.VGOLIAX IBEX, Capra sihcrica. 1923. 






nk ^ 


, .tTl«f-^ 

-< J . "<. 


'«£. -A^ 




Futvtre study of petroglj^ihs of this type by otir archaeologist, Mr. Nelson, 
confirmed the geologists' opinion that they were made by a pre-Mongol people. 


The next day a heavy rainstorm kept us all in camp. Just at sunset we 
saw a dark blotch on the plain. Our Mongols, with their wonderful eyes, 
assured us that it was our caravan, and before long, with glasses, we could 

Figure 9. — Drawings cut upon a basalt block at Artsa Bogdo. From "Geology of Mongolia.' 

distinguish the American flag ; an hour later we saw a single rider loping up the 
slope and Merin came into camp. When I complimented him upon his prompt 
arrival he remarked: "You said to come in three days and so we did." But 
the distance was twenty miles greater than we had estimated when I told him 

The baby wild ass which had been with us since we captured it on July 
6 at Tsagan Nor, died August 18. We had not been able to wean it and were 


never able to get a sufficient quantity of goat's milk; at Artsa Bogdo there 
was cow's milk in abundance and the little animal developed severe diarrhoea, 
resulting in death. "Buckshot," the Chinese boy who tended it, was the only 
member of the Expedition whom it would tolerate. 

It was important to start the caravan immediately toward Sair Usu, a 
well on the old Chinese official road which we hoped to reach on our return 
journey to Kalgan. The camels could travel overland regardless of the terrain 
but just how we were to reach Sair Usu was uncertain. We took food and gaso- 
line enough for three weeks, and the caravan left August 20. 


Granger decided that he could prospect the Artsa Bogdo basin more suc- 
cessfully from a base down on the plain, so he left us to camp at a spring 
fifteen miles to the north. Berkey and Morris wished to study the Artsa 
Bogdo uplift and the region directly to the south. They could best do this by 
riding through the moimtain itself on camels and ponies. With a guide and 
animals hired from the natives, they left on the nineteenth ; thus the Expedi- 
tion was again divided into four units. My own task was to collect a group of 
ibex and sheep for the Hall of Asiatic Life, and see that the taxidermists 
obtained a representative collection of the small mammalian fauna. 

I hunted the mountains behind camp for several days and discovered that 
none but females and young of either sheep or ibex were to be found there. 
My native guide assured me that big rams of both species were abundant in 
the high peaks of the western part of the range. Consequently, I engaged a 
Mongol lama as guide and rode away for a four-day shoot. Our method was 
to fill two canvas water-bags and oiu- canteens, ride to the highest peaks, and 
sleep wherever we found game. As there was no water on the mountain sum- 
mits, we rode dowTi to some stream in a valley bottom, after the early morning 
shoot, to water the horses and ourselves. 

I had ridden the Mongol ponies for months over the desert and grasslands, 
but this was my first experience with the animals in mountain climbing. They 
amazed me by their strength, endurance and sure-footedness on slopes so steep 
as to be seemingly impassable for any animal as large as a pony. The western 
end of Artsa Bogdo is cut by two or three deep gorges with partly dry stream- 
beds in the bottoms. The tremendous erosion which has taken place indicates 
that at some time in the past there was a period of moistvire, when heavy rain- 
falls enabled the rivers to do their work. One of the gorges is at least seven 
hundred feet deep for most of its way. 

My Mongol guide proved to be one the best himters whom I have ever 
had. He had hunted all his life, until three years before, when he had become 
very ill. Believing that he was about to die, he made a vow to Buddha that if 


he recovered he would become a lama. As a result he was supposed not to 
take life, but his natural passion for hunting was too much for him, and fre- 
quently he stole off to use the old flintlock gun which he kept concealed in a 
small rock-cave. He had an almost uncanny knowledge of what the game 
wotild do, and just as great skiU in stallving. He never had seen a high-pow- 
ered rifle operate, and it so happened that, the first time he was with me, I 
killed two sheep across a ravine at a range of four hundred yards. He was 
literally speechless with astonishment, and more amazed, if possible, when he 
saw the tiny 6 mm. bullet and then examined the hole it had torn in the sheep's 
side. As a result he believed that I could kill an animal as far as I could see, 
and it took some time to convince him that there was a limit to the "range of 
my rifle. 

I found that the Mongols were quite right in saying that the big rams of 
both ibex and sheep were among the western peaks. In the four days that I 
hunted there I saw about one hundred and fifty ibex and half as many sheep. 
As far as I know we were the first white men to hunt the moiintain, but the 
animals were by no means vmwary, for the Mongols continually pursue them 
during the winter. The following year McKenzie Young and I spent three 
weeks shooting on this same moimtain, and visited it again in 1925 with six 
of the Expedition members. 

Neither the male nor female ibex is ever very far away from rocks. We 
often fotmd bighorn sheep in ibex country, but never saw ibex out on the rolling 
grassy hills which sheep love. During the simimer the ibex and sheep both 
feed from the first gray light of dawn until the sun is high ; then they sleep in 
the shade nntjl about two hours before dark, when they begin to move about 
again, keeping well in the shadows. A saddle or a depression on a ridge is a 
favorite sleeping place because there the wind reaches them from every side. 
They depend much less upon their sight or hearing than upon smell to protect 
them from enemies. They seem to know by instinct those spots "where the 
baffling mountain eddies chop and change." 

The dark brown of the summer coat of the ibex is a marvelous protective 
coloring. Often I have siurveyed the side of a rock-cliff with my field-glasses, 
scrutinizing every inch, yet nothing would be visible until a movement would 
show that what I thought was a rock was in reality an ibex. The females are 
much more diffictdt to see than the males, because the long curved horns of the 
latter often will show in silhouette when the body is concealed. 

Sheep and ibex are seldom off guard. When a herd is feeding or asleep two 
or three sentinels are posted. One morning, Young and I watched forty ibex 
graze up an almost perpendicular moimtain side until the last foot of shade had 
disappeared, and then dispose themselves comfortably among the rocks. 
They were plainly visible while they were standing, but one by one they faded 


from sight and seemed literally to sink into the grovmd. Only two bucks were 
left. They climbed lazily to the highest peak and took stations side by side 
facing in opposite directions. One surveyed the vast complex of mountains 
to the south; the other gazed over the plain, which stretched away like a calm 
sea. For two hours they stood motionless, living statues silhouetted against the 
sky. Then, at the same moment, they left the sentinel post and lay down. 
We watched a similar performance with a herd of sheep. In that instance one 
ram remained on guard. He climbed to the topmost pinnacle of rock on the 
highest hill nearby and stood as if carved in stone. Eventually he left and lay 
down himself, leaving the herd unguarded. 

One hears a great deal about the enormous sheep and ibex heads of the 
Altai and Tien Shan. There seems to be no question that the Tien Shan ibex 
do produce larger horns than those of the Altai. Also, larger bighorn sheep 
Uve in the western Altai than in the eastern part of the same range where we 
hunted. We always found the largest ibex and sheep in the highest and most 
rugged mountains, and, of course, the eastern Altai do not compare in height 
with the western part of the range. 

We shot twenty-five specimens on the Artsa Bogdo in 1922, 1923, and 
during that time I counted several hundred ibex and half as many sheep, yet 
I did not see more than ten sheep with strikingly large horns. I saw only three 
ibex with horns that would have measured fifty inches or more. One day I 
killed two ibex from a herd of nineteen bucks; the horns of one measured thirty- 
seven inches and of the other forty-one inches. I inspected the herd careftilly 
before shooting and am sure that none of the animals carried horns larger than 
those of the ones killed. Lieutenants Butler and Robinson shot an ibex with 
forty-one-inch horns, as did Doctor Loucks in 1925. I shovild say that the 
usual horn length for the Artsa Bogdo ibex is about thirty-six inches. The horn 
average for the sheep on this mountain is about forty-five inches in length with 
a basal circumference of eighteen inches. As I have remarked, I did see about ten 
sheep with horns which must have been close to sixty inches in length and 
twenty inches in circumference at the base. Unfortimately, none of us got a 
sheep with horns longer than fifty inches; neither did we see any large horns 
lying on the mountain slopes. The Gurbun Saikhan, the next eastern moun- 
tain imit, is still lower than Artsa Bogdo and consists of grassy hiUs and 
plateaus. It is ideal sheep ground and there are many argali there, but we 
saw no really large ones, and the natives told me that big rams were rare. 
Farther east the Altai Mountains dwindle into isolated and stiU lower units, 
almost all of which are inhabited by sheep. I am convinced that the big heads 
are to be fovmd in abundance only in the western Altai among the higher 

J. H. Miller remarks: 














"Though more than one Russian explorer mentions having met sheep in 
the eastern Altai, it is to Sir Francis Younghusband that we are indebted for 
by far the most interesting information on this subject. In his remarkable 
journey of 1887 across the Gobi, from Peking to Hami, he struck the great 
Altai at its most easterly extremity in longitude 100° East. He estimated the 
height of the range, even at its terminal portion, as 9,000 feet above sea-level, 
and the natives reported grassy plateaux in the centre. These two combina- 
tions sound suitable for amnion. Though Sir Francis did not visit these high 
plateaux, where the sheep would have been at that season (Jtily), yet on the 
outlying southern foothills horns were found lying on the ground which, from 
their great girth of 19 inches and general shape, undoubtedly belonged to 
Ovis amnion typica."^ 

The most eastern point at which we hunted sheep was Jichi Ola, a low 
isolated mountain at Longitude E. 106°, Latitude N. 43°3i'. This is nfearly 
five himdred miles east of Youmghusband's records. From the motor cars we 
saw sheep at several other isolated mountains which were at the very extremity 
of the main Altai chain. 

We saw demonstrated beyond a doubt the fact that sheep will cross con- 
siderable stretches of open plain to reach a distant mountain. About four 
miles north of Jichi Ola there was a small, low mountain standing alone in the 
desert. One day Granger discovered a herd of ewes led by a yoimg ram far 
out on the plain proceeding leisurely toward the isolated mountain. He fol- 
lowed them in his car and found that the highest speed which they could reach 
was twenty-five miles an hour. They quickly became exhausted and gathered 
into a compact group. He stopped a hundred yards from them and while he 
was trying to get his camera the sheep suddenly separated and dashed away. 
They were doubtless making this pilgrimage to escape from the higher mountain 
where we had been shooting for two days. 

McKenzie Young and I were hunting at the extreme eastern end of Artsa 
Bogdo in 1923. We had been following a large herd of sheep and were surprised 
to see them leave the moimtains and dash across the desert to the southeast. 
Another herd abandoned the plateau to go out upon the plain, but we met them 
again returning to the foothills several hours later. The Mongols at Artsa 
Bogdo told us that they often saw sheep crossing ten or fifteen miles of desert 
from one motintain to another. Typical Ovis ammon ranges as far east as Lon- 
gitude E. 110°, for I discovered a skull with good horns lying in the grass near 
Gatun Bologai, a well-known spring on the Kobdo-Kweihwating trail. This 
was half a mile from a low, ragged granite outcrop about five hundred feet high. 
Between it and Jichi Ola there are isolated mountains, not more than five or ten 

' Caxruthers, Douglas. 1914. "Unknown Mongolia." With three chapters on sport, by J. H. Miller. 
Hutchinson & Co., London, p. 346. 


miles apart, which probably act as "stepping stones" for the sheep. According 
to the Mongols these lowland crossings take place during the winter months. 
I believe that Longitude E. i io° is the utmost eastern range of true Oiis amnion. 
Several himdred miles of arid flat desert separate this point from the moimtains 
of the Sino- Mongolian frontier, where another related species, Ovis comosa, is 
found. Dr. Glover Allen, after studj-ing specimens of Oins comosa, concludes 
that it does not differ specifically from the true Chis amnion. 

Miller speaks of frequently seeing wolves in the western Altai where he 
hvmted ammon, but we did not see a single wolf or any traces of them at Artsa 
Bogdo. I have already remarked how scarce they are upon the desert, but we 
had expected to find them in the moimtains. Except for man, the sheep and 
ibex have no important enemies in this region, imless the lammergeyer may be 
counted as one. Probably this great vulture does get a good number of lambs 
but I doubt whether it attacks adult sheep or ibex. 


Both the avian and the mammalian faunas have followed this far eastern 
extension of the Altai from the high moimtains of the west. It is also inhabited 
by several species of birds and mammals common to the mountains of north 
China and MongoHa, both north and south of the desert. We never saw the 
lammergeyer except in the Altai; also, the snow-cock, Tetraogallus, was an 
Altai bird not observed elsewhere. The chukars, quail, bearded partridges and 
red-billed chough, all birds of the north China and Mongolian border moun- 
tains, were abundant at Artsa Bogdo. The mole-rat, Ellobius larvatus, of the 
Altai we caught at Sain Noin as well as at Artsa Bogdo. The cony, Ochotona 
pallasi, was not peculiar to Artsa Bogdo, as the same species was collected at 
Uskuk, Gun Burte and Tsetsenwang. The spermophile, Citellus obscurus, the 
marmot, Marmota hobak siberica, and other small mammals were also found at 
other localities in the Gobi. 

Marmots were fairly abundant on the high plateau of the mountains, 
although there was no sign of them on the lower slopes. This was unlike any 
other spot where I have encountered marmots; usually they prefer the foot- 
hills to the high mountains. I cannot believe that it was because they had 
been driven away by Mongols, for in that case we wotdd have seen abandoned 
burrows. Conies were very numerous, and in the meadow-like plateau between 
the peaks their little haycocks dotted the slopes; some of them were as much as 
two feet in height. 

The quail and chukar partridges were extraordinarily abundant, and as 
the yoimg were just able to fly well they gave us good sport and delicious food. 
When we returned the next j^ear (1923) there was hardly a bird to be found; 
in three weeks' shooting I saw only two quail and one chukar on the mountain. 


What happened to almost exterminate them I cannot imagine. The natives 
said that the winter had not been unusually severe. Both in 1922 and 1923 
there were many Ovis ammoji on Artsa Bogdo, but in 1925 our shooting parties 
saw only two or three sheep, although the ibex were as numerous as in previous 

The snow-cock was a new bird to me, for it is an inhabitant of the Him- 
alayas and the western Altai. We found the birds only among the rocks on 
the highest peaks. They utter a weird, rapid-fire call when sailing across a 
ravine from peak to peak, a call which seems to fit the wild movmtains among 
which they live. I saw them only when I was hunting ibex and could not 
shoot for fear of disturbing larger game, but my Mongol, Tserin, was very 
anxious to have me kill one or two for him. He said that the Chinese esteem 
the meat for medicine, as well as the dried blood. I said, "Tserin, do you really 
think the meat would ctu-e you if you were ill?" 

"Probably not," said he, "but the Chinese will pay ten dollars for it and 
that is good enough for me!" 

Artsa Bogdo takes its name from the low juniper, Juniperus sabina, which 
the Mongols call artsa. This plant spreads over the rocks and grows in masses 
in the bottoms of the depressions and ravines among the moimtain heights. 
The bighorn sheep often sleep in the artsa and we followed their example with 
great comfort. The artsa is confined to the western end of the mountain. I 
saw none of it on the grassy hills of the central and eastern parts, where there 
are few rock outcrops. 


On August 28, Berkey and Morris retiuned from their trip across the 
moimtains. They had traversed the desert to the south of Artsa Bogdo, 
reached the western end of the Gurbun Saikhan, and returned by a different 
route. It had been a trying expedition, for both men were pretty well exhausted 
from their strenuous work at Baga Bogdo, and the weather was hotter than 
any we had experienced earlier in the season. Nevertheless it had been 
profitable, and after a few days' rest they were fit again. 

Beside making geological observations they discovered some palaeolithic 
artifacts of chalcedony near a spring on the south side of Artsa Bogdo. These 
were the first indications that we had had of any very ancient hioman occupation 
of the region. They also fotmd a few fossil bones which Granger decided were 
dinosaurian, thus identifying Cretaceous strata south of the Altai Mountains. 


Perhaps most important of all the finds by Berkey and Morris was a single 
piece of red limestone containing invertebrate fossils. Dr. A. W. Grabau, who 


examined it, said that the fossil fragments, which include crinoids and corals, 
suggest either late Devonian or Dinantian age. It was particvdarly interesting 
as being the first evidence of Palaeozoic strata found in Mongolia (see Vol. 
IV, this series, p. 22). 


Regarding the general structvire of Artsa Bogdo, Professors Berkey and 
Morris, 1927, have said: 

"The representatives of the Altai system in the mid-Gobi owe their 
uplift and relief to block faulting. A row of narrow blocks, whose longer axes 
lie east and west, has been elevated and tilted. The whole system dies out 
toward the east, each block being lower than its western neighbor. It is an 
incidental matter that the rocks involved have exceedingly complicated struc- 
tural features and exhibit close folding nearly parallel to the elongation of the 

"The Artsa Bogdo differs from the Baga Bogdo vmit in that the dominant 
rock formations are crystalline metamorphic representatives of the ancient 
series, whereas Baga Bogdo exposes an immense area of granite, flanked by 
metamorphic series. This difference, we judge, is in large part the result of a 
different amount of elevation and erosion. Most of the stratified and meta- 
morphic cover of Baga Bogdo has been stripped from the imderlying bathy- 
lithic granite, whereas from Artsa Bogdo the stripping is not so complete and 
the usual cover of ancient metamorphosed strata is preserved throughout most 
of the range. "^ 


We were entranced with the open park-like tracts in the central and 
eastern part of the Artsa Bogdo imit. Long meadows, as smooth as a putting- 
green, swept down from rounded grassy summits into gently undulating val- 
leys like the billows of a great ocean swell. From them we could look across 
the blue haze of desert far below us to the dim outlines of a blood-red mesa 
forty miles to the north. We did not know imtil after we had left the moim- 
tain that this superb mesa stood in the center of a basin, or desert-hollow, 
which Granger had discovered and where he was then at work. We named 
it the Oshih Basin. The fossiliferous exposures were of Lower Cretaceous 


On August 30, we left Artsa Bogdo and moved to a spring a few himdred 
yards from the trail. Granger had camped there for several days and had sent 
word that he would despatch a man to guide the geologists to his new fossil- 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geolog>- of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. v-xxxi, 1-475. 


bearing basin which he described as well worth investigation. Since I had much 
to do in preparation for the start of oiir homeward journey on September i, 
I remained in camp with Shackelford. 

The geologists and Granger returned the next day. They were in the high- 
est spirits and had several boxes of fossils. Granger reported the basin to be a 
wonderful spectacle of erosion; an amazing chaos of canyons, ravines and 
giillies, in the midst of which stands the red mesa capped with jet-black lava. 
There were remains of an interesting ancient wall made from lava blocks, 
probably pre-Mongol, and an abiuadance of fossils. 

Granger brought in two skeletons of small dinosaurs, one of which was 
beautifully preserved, including an almost perfect sktill and jaws. The speci- 
men was discovered by Wang, the Chinese chatiffeur who had foimd the first 
bones of the Baluchitherium at Loh. Professor Osbom named this dinosaur 
Psittacosaurus mongolie7isis. It is allied to, or possibly identical with, Pro- 
tiguanodon from the Ondai Sair formation of Uskuk. Professor Osbom 

"These two types resemble each other in so many characters that they 
obviously belong to a distinct family of iguanodonts to which the name Psitta- 
cosauridae has been applied. These short-skulled iguanodonts derive their 
family name Psittacosauridas from the very deep parrot-like beak, with smaU 
nostrils located at the top of the very deep maxilla."^ 

Granger made another important discovery, consisting of two charac- 
teristic teeth of a large sauropod dinosaur. Professor Osbom, 1924, names the 
sauropod Asiatosaurus mongoliensis and concluded that the teeth resemble 
Camarasaurus. Also, two other teeth, of a carnivorous dinosaur designated 
Prodeinodon mongoliense, were foimd in the same formation. The discovery of 
both Sauropoda and Theropoda added two more great groups to the list of 
Mongolian prehistoric fauna which our single season's work was bringing to 
formidable proportions. The formation was named Oshih (Ashile), and it 
includes the oldest of the later sediments lying upon the complex oldrock 
floor of Mongolia. Since the dinosaurs Psittacosatirus and Protiguanodon 
found in the Oshih and Ondai Sair formations are possibly identical or at least 
very closely related, they indicate that the two formations are of approximately 
the same age. This may be said to be near the beginning of the Lower Cre- 

Granger had been away for ten days and had been so busy removing speci- 
mens that he had had little time to prospect the great depression. The basin 
gave promise of being a most important fossil locality, but, strangely enough, 

' Osbom, Henry Fairfield. 1924. "Psittacosaurus and Protiguanodon: Two Lower Cretaceous Iguano- 
donts from Mongolia," Amer. Mus. Novitatcs, No. 127, p. i ; Preliminary Reports, Central Asiatic Expeditions, 
Amer. Mus. Nat. Hist., I, No. 25, pp. 1-16, 1926. 


subsequent work showed that it was the only one of our 1922 localities which 
did not fulfil expectations. 

I had set September i as the date of our departure, for we had many hun- 
dreds of miles of unknown country to traverse and there were abimdant signs 
of approaching winter. Baga Bogdo had already assumed a new crown of 
snow and we might expect a blizzard at any time. A hea\^ fall of snow would 
be disastrous, for, even if it did not remain long, it would make traveling diffi- 
cult for our motors wherever there was soft grotmd. Therefore, we marked 
the Oshih Basin as a field for further investigation and prepared for the home- 
ward trip with light hearts. 



Our last camp opposite Artsa Bogdo moimtain was on the ancient Kobdo- 
Kweihwating trail, ten miles north of the mountain. We decided to follow the 
trail eastward to the vicinity of Ulan Nor and then try to strike northward to 
the Uliassutai-Sair Usu road, which we believed cotild be traveled safely into 
Kalgan. If this route proved impracticable we should have to find some 
other way to Kalgan. 

The crisp cold air, the uncertainty of travel and the prospect of new and 
interesting country put everyone in high spirits. Nowhere in the world have I 
known air so invigorating as that of Mongolia in the autvmin. The altitude, 
nearly five thousand feet, and the desert dryness make even the warmest days 
of summer bearable, and the really hot weather lasts only about a month. 
The final two weeks of August and the first part of September bring an autumn 
sharpness which is as delightftd as it is stimulating. Oxir bodies siu-ged with 
vigor and the most strenuous work seemed not to tire us. It is easy to believe 
that the proper theater of evolution for the htiman race must have been on 
uplands with a summer climate similar to that of the present-day MongoHa, 
rather than in the tropics with their debilitating heat. 

We left on Friday, September first. Driving eastward with the dark 
mass of Artsa Bogdo still in sight, we made excellent progress over a hard trail. 
The cars were heavily loaded, for we had filled all available space with gasoline 
and Granger's collecting in the Oshih Basin had added several heavy boxes of 
fossils. Thirty miles down the trail an accident to the clutch of one car delayed 
us nearly a day, but the geologists utilized the time to make closer investiga- 
tions of the geological features of the region. Continmng southeastward we 
found the trail for the most part good, except in three or foiu- places where it 
was amazingly bad. Fifty-three miles from Artsa Bogdo we found a shallow 
basin filled with "nigger-heads," where the water-table was close to the surface 
and the red soil soft and wet. It was difficult terrain, and the heavier cars were 
often mired so badly that we had to unload and dig them out. For many miles 



we had seen no Mongols. We hoped to find natives who could give us informa- 
tion of any caravan trails leading northward, but it was not iintil we were 
sixty-five miles east, opposite the low-lying mass of the Gurbun Saikhan, that 
we saw a group of yurts some distance off the trail. 


My car was far in advance of the others and I asked Shackelford to stop 
the fleet while I ran over to the yurts for a conference with the inmates. Dior- 
ing the time that I was gone he wandered off a few hundred yards to inspect 
some peculiar blocks of earth which had attracted his attention north of the 
trail. From them he walked a little farther and soon foimd that he was stand- 
ing on the edge of a vast basin, looking down upon a chaos of ravines and gul- 
lies cut deep into red sandstone. He made his way down the steep slope with 
the thought that he would spend ten minutes searching for fossils and, if none 
were foimd, return to the trail. Almost as though led by an invisible hand he 
walked straight to a small pinnacle of rock on the top of which rested a white 
fossil bone. Below it the soft sandstone had weathered away, leaving it bal- 
anced ready to be plucked off. 

Shackelford picked the "fruit" and returned to the cars, just as I arrived. 
Granger examined the specimen with keen interest. It was a skull, obviously 
reptilian, but unlike any with which he was familiar. All of us were puzzled. 
Granger and Gregory named it Protoceratops andrewsi in 1923. Shackelford 
reported that he had seen other bones, and it was evident that we must investi- 
gate the deposit. 

We could see a small pond on the bottom of the basin, and two Mongol 
soldiers whom I had brought with me from the yurts said that there was a 
well near-by. We decided to camp where we were and utilize the three or 
four remaining hoiu-s of daylight to prospect the exposure. The tents were 
pitched on the very edge of the escarpment, and every available man of the 
expedition, native and foreign, went down into the badlands. Quantities of 
white bones were exposed in the red sandstone, and at dark we had a sizable 
collection. However, Shackelford's skull still remained the best specimen, mth 
the possible exception of the skull and jaws of a small reptile, fotmd by Berkey. 

Granger brought in, among the other things, a part of an eggshell which 
we supposed was that of a fossil bird, but which subsequently was recognized 
as dinosaurian. It was evident that the formation was Cretaceous and very 
rich in fossils, but at that time we could do no more than mark it as one of the 
localities for future work. We could hardly suspect that we should later con- 
sider it the most important deposit in Asia, if not in the entire world. Sub- 
sequently it was formally designated as the Djadochta formation of Shabarakh 







Cretaceous iK'ds of Shabarakh Usu. 




This is one of the most picturesque spots that I have ever seen. From 
our tents, we looked down into a vast pink basin, studded with giant buttes 
Hke strange beasts, carved from sandstone. One of them we named the 
"dinosaur," for it resembles a huge Brontosminis sitting on its haunches. There 
appear to be medieval castles with spires and turrets, brick-red in the evening 
light, colossal gateways, walls and ramparts. Caverns run deep into the rock 
and a labyrinth of ravines and gorges studded with fossil bones make a para- 
dise for the palaeontologist. One great sculptured wall we named the "Flaming 
Cliffs," for when seen in early morning or late afternoon sunlight it seemed to 
be a mass of glowing fire. On the floor of the basin, to the north, is an area of 
old dead sand-dunes covered by a miniature forest of stimted trees, which we 
first supposed were tamarisk, but which have been identified by Dr. Alfred 
Rehder, 1927, as Salicornia herbacea. 

To the south, the rolling plain sweeps back to the Gurbun Saikhan, the 
last of the prominent uplifts of the Altai system. Gurbun Saikhan means 
"the Three Good Ones," and we found later that the name is well bestowed, 
for it is a fair land of rolling meadows and long grass among rounded peaks. 
Of the group. Baron Saikhan stands at the western end, Dvmde Saikhan in 
the middle and the Dzun Saikhan at the east. They are essentially ranges of 
mountains separated by wide valleys, so that they appear as three major 
units. Like the other members of the Altai system they exhibit fault-block 
featvires in their construction. 


The yiirt which I had visited for information was a customs station 
between the province of Sain Noin Khan and that of Tushetu Klhan, which 
we had just entered. The soldiers were very decent fellows and seemed 
unusually intelligent. They produced a gxiide who was supposed to know 
of a small trail which led northward through the basin to the Sair Usu trail. 
We hired the man and he slept that night with our Mongols. The next 
morning he led us down an almost impassable slope to the eastern side of 
the basin floor, which was thickly covered with sagebrush and playa-tufts. 
It was bad going, but at last we foimd a faint trail and continued on it for 
ten miles until it turned down into the Ongin River valley, which we were 

The terrain became more and more composed of soft sand covered with a 
heavy growth of low bushes and at last the cars could go no further. We were 
much discouraged for so much gasoline had been constuned that we doubted 
if enough were left to take us to Sair Usu should it be necessary to retrace our 
way and make another attempt. 



While I was debating the question, Granger and Berkey prospected ahead 
along the very edge of the bluff which dropped off to the river valley. They 
found less sand there and an absence of woody growth, and we decided to push 
on. Within two miles the terrain became harder, and we found that the vast 
plain had been swept clear of sand, leaving a smooth gravel surface. Hardly 
a trace of vegetation broke the gray expanse, which we named the "Htmdred- 
Mile Tennis Court." 

This was one of the smoothest and longest-continued level stretches seen 
anywhere in Mongolia. For about sixty miles the difference in elevation 
amounted to only five hundred and fifty feet, and there were areas where for 
ten miles the total variation was not more than one hundred feet, with no local 
irregularity of more than ten feet. Apparently the area was as level as a floor, 
but the whole plain rose gently northward at the rate of about ten feet per mile. 
During ahnost the entire northward run we followed the Ongin River, on the 
eastern side of its valley. It was a wide trench carved out of the plain to a 
depth of fifty feet, and as the shallow river was split into numerous branches 
there was much green grass and many bushes in the valley bottom. 

The last forty miles of the day's run were as delightful as the first part 
had been discouraging, and we sped along at twenty-five miles an hoiir. One 
or two small groups of yurts were pitched on the edge of the river valley, but 
we passed them without a pause, leaving frightened and wondering Mongols to 
stare after the apparitions that had suddenly appeared in their midst. Doubt- 
less none of them had even heard of an automobile, and I would give much to 
know what were their thoughts about us. 


We camped that night on a green meadow within the valley near the 
lamasery "Ongin Gol-in-Sumu." It was a large establishment, with hundreds 
of lamas and novitiates, who literally swarmed about our tents and cars. They 
seemed of a rather better class than the usual priests and some of them had 
fairly intelligent faces. Also, they appeared to be under strict discipline, and 
the yovingsters were sent scurrj'ing back to their lessons when their instructors 
deemed that they had seen enough. 

The bviildings themselves were not so picturesque upon nearer view, for 
many of them were of gray brick. A large temple of wood was being con- 
structed by Chinese carpenters, who do all the temple building in Mongolia. 
The Mongols never attempt work of this sort themselves. 


The lamas told us that our caravan had passed that way ten days earlier, 


and that the Sair Usu trail was only thirty or forty miles away ; also they gave 
us the cheering prospect of good going. We did find the trail hard and smooth, 
although it soon turned away from the Ongin River valley and climbed up and 
down steep hills composed of Jurassic sandstones and conglomerates. The 
country was very desert-like and we saw only a few yurts, in some of the larger 
valleys where there was a little vegetation. We reached the Sair Usu trail 
just forty-seven miles from the Ongin Gol lamasery. It seemed certain now 
that there woiild be no serious difficulty in making the run to Kalgan, for the 
trail was wide and well-traveled. 

As I have remarked before, this was the road which officials used in mak- 
ing visits from China to Urga, Uliassutai, Kobdo and other parts of Mongolia 
during the Manchu regime. We found wells at an average of ten miles apart, 
and the water was clear and cold although slightly alkaline. At several places 
the ruins of small Chinese mud houses were stiU to be seen; doubtless these 
were official stopping-places when the road was in use during the Empire. 

We had hoped to find other sedimentary basins in this part of the Gobi, 
but it soon became evident that we were in an area of igneous rock formations 
of Jurassic age. Some of it was exceedingly complex and the geologists had a 
very busy time solving the puzzles, as we made only infrequent stops. We 
covered one hundred and twelve miles that day — the longest rvm of the summer. 
As I look back upon it and realize what the luicomplaining Berkey and Morris 
endured, I am decidedly contrite. They have remarked, 1927, concerning it: 

"Nevertheless, the geologic route section was kept fairly true. It is a ter- 
rific task to unravel more than a hundred miles of such structtire in a working 
day. The geologist finds himself completely exhausted, and even dreams of 
tmderground structure, of strata and fossils, bathyliths, roof-pendants and 
ancient metamorphic revolutions, until the sun rises again on another day of 
geologic puzzles. One must try it to appreciate in full the effort that must be 
made to keep the story straight."' 


The remainder of the run was vmeventfvil. In mid-afternoon we sighted a 
half-dozen mud huts, partly in ruins, and a small temple and ruins of a larger 
one, in a sandy basin beside two wells. Such was Sair Usu. In the days of the 
Empire it must have been a station of considerable importance, for it is at the 
intersections of main trails to Urga, UUassutai and Kobdo. A little to one 
side of the nearest weU stood the blue tent of our caravan and the long lines of 
boxes with the American flag in the center. Merin and his Mongols welcomed 
us joyously; the day was September 5 and we had kept oiu- appointment. 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. v-xxxi, 1-475. 


Merin had arrived late the previous evening, without incident save the loss 
of one camel. We had seen its skinned carcass beside the trail, for the Mongols 
always exhibit the skin as evidence that the animal has really died and has not 
been sold. I always allowed the caravan leader to keep the money obtained 
from the sale of the camel and sheep skins, and the camel's wool which they 
collected dining the summer, except that used by the palaeontologists for pack- 
ing fossils. 

I spent all next day taking out the necessary food and gasoline for the 
run of five hundred and thirty-six miles to Kalgan, also, in transferring to the 
caravan the specimens collected and every item of equipment that could be 
spared, for we should not see the camels again. Late the previous night a big 
Chinese caravan had come in and camped a short distance away. The members 
of this caravan were on their way to Kobdo and Chinese Turkestan, carrying 
cloth, tea, tobacco and boots, to be bartered for wool, hides, furs and ponies. 
The visitors could give us little news of China, since they had been away two 
months, but they did tell us about the trail. 


We left Sair Usu on September 7. I gave IMerin final instructions to go 
straight down the trail to Kalgan and to reach there not later than October 20. 
Actually, he arrived ten days earlier than that date with all his specimens and 
camels in good condition. The trail took us southeast, up and down a series of 
low hills which afforded rock outcrops that gave Berkey and Morris a very 
busy time. Eleven miles from Sair Usu they discovered a cliff of fossiliferous 
limestone just south of the road, and a hasty inspection revealed a fauna indi- 
cating Palaeozoic age. Subsequently it was determined as Dinantian (Missis- 
sippian) . The geologists were able to collect only a few specimens and to note 
the locality as an important one for futiore investigations, which Morris made 
the following year. This was the first time that determinable Palaeozoic strata 
were discovered in place. Only one other piece of Palaeozoic rock, a fragment 
from the alluvial slopes at the foot of the Gurbun Saikhan, had been found up 
to this time. 


That day we saw once more grim evidence that for the preceding three 
years a human life in Mongolia had been worth less than that of a sheep. At 
one place, Shackelford and I, in the leading car, drove over to two yurts to 
inquire about the road. We saw three men leap on ponies and ride to the hills, 
leaving four women. Two were very old, one was about fifty and the fourth 
was a really beautiful girl of eighteen or twenty. They had spread a clean 
white felt before the >airt and were lined up, trembling and kowtowing. As 


we stopped the car a few feet away, the girl ran to get another felt, and one of 
the women rushed inside to bring milk and tea. In a few moments our Mongol 
explained that we were Americans and would not hurt them. They had never 
heard of America, nor of any white men except Russians. When I gave them a 
few trinkets they were pitifully pleased. They clung to one another, crying, 
and explained that they had expected to be killed instantly. They had wanted 
to run away, but the men had taken all the horses, and we came so fast that 
they cotdd not hide. A short time later we stopped at another yurt, and one 
of the two women there had an attack of violent nausea, from sheer fright, 
when she saw Granger and Colgate. There were a large number of yurts in 
the less arid parts of this region, and several small temples. Such places have 
only a few resident caretakers, but lamas come to these temples for two or three 
months during each year to conduct services. 


At one bad sand-wash there was an accident to the pinion gear of one of 
the heavily loaded cars, which delayed us for an afternoon and morning, but a 
new part was substituted and we went merrily forward again. Hard work 
developed almost immediately, because the trail led us into a dry, sandy river- 
bed between high cliffs, where shoveling was the order of the day. Our eyes 
were gladdened, however, by the sight of several good-sized elm trees grow- 
ing in the dry stream bottom. Here their roots penetrated deeply enough to 
find water even in the seasons of greatest drought. None of us realized how 
much we missed trees until we heard the soothing rustle of the wind among 
the leaves. 

After the bad going we came out into a great depression where the trail 
was hard and smooth. This basin was one of the most arid stretches that 
I have ever seen, a rolling gravel floor with only the scantiest vegetation and 
hardly a trace of animal life. The deathlike monotony of the bare, gray 
desert was most depressing, and the expedition lapsed into silence. For forty 
miles there was no sign of a human being; then we discovered a Chinese cara- 
van from Kweihwating en route to Uliassutai. The men of the caravan told us 
of a well some ten miles farther on, but if later we had not happened to meet 
two Mongols on horseback, we never should have found it. The Mongols 
never had seen an automobile and were badly frightened at first, but, unlike 
most natives, stood their ground while we drove up to them. I asked one to 
ride in the car and guide us to the well. It was amusing to watch the play of 
emotions reflected in his face; first astonishment and negation, then hesitation, 
next extreme desire, and finally surpassing courage as he acquiesced. 

His companion led his pony to the well, which was two miles off the trail. 
Two yurts were near-by and I sent one of our Mongols to buy a sheep. He 


found only one old man, frightened nearly to death, who said that all the 
others had fled to the hills at our approach and had driven their sheep with 
them. Even though we left presents with the old man, the Mongols returned 
during the night and quietly packed up their yvirts ; when we awoke the only 
indications that they had ever been there were the dead embers of an argtd 

After a morning run on the same gravel plain, we came to a sand-swept 
surface, where with every mile the cars labored more heavily. They could 
barely make headway on the trail, which was only faintly indicated in the loose 
sand, but at last we descended a moderate slope into a great basin where the 
terrain was again clean hard gravel. 


A line of cliffs bulked faintly against the sky, and as we neared it, red, 
gray and yellow bands showed on its bold face. My car was far in advance of 
the fleet and I drew up at the base of the bluff to await the others. I told the 
servants to prepare tiflSn while the scientists gave the exposvu-e a hasty inspec- 
tion. Berkey, Morris and I prospected the red basal sediments unsuccessfully, 
but Shackelford, with the instinct of a pointer dog for game, worked along the 
overlying gray and yellow sands. He returned with his pockets full of bones 
and teeth, among which carnivores and rhinoceros were easily identifiable. 
Granger had stopped at a low rovmded knoll beside the trail, and brought 
fragments of titanothere as his contribution. 

It was quite evident that we must give the exposure a more thorough 
inspection, and I told the boys to make camp. The bluff extends east and west 
for many miles, but the direction changes abruptly to almost north and 
south at the prominent comer where the trail passes and we had camped. At 
this point the actual face of the escarpment measures three hundred feet, 
and the slopes below it add another two hundred feet. It stands out like a 
cape on a seacoast. The impression is strengthened when one climbs to the 
top and looks over the vast basin below. On the very point of the bluff the 
Mongols had built a large obo, or prayer monument. Many obos are mere piles 
of stones on the simimit of a hill near the trail, but others are more elabo- 
rately constructed. Each stone means a prayer to Buddha, and the Mongols 
never fail to add a contribution to the pile when passing such a monument. 

We foimd that the bluff was named Ardyn Obo and was well known to 
the natives for many miles. It was appropriate that this name should be 
given to the formation. The obo itself was solidly built and elaborately deco- 
rated with silk hatas, torn bits of cloth, sticks and roughly shaped wooden 
spears and knives. The Mongol word ardyn} means "jewels," referring to 

' Spelled Irdin for the locality south of Iren Dabasu. 





the highly polished pebbles of quartz and chalcedony which are found in the 
upper sandstones of the formation. 


The face of the bluff has virtually no vegetation, and the strata are well 
exposed. They consist essentially of three members, red clays below, gray clays 
and yellow sands in the middle and cross-bedded sands and gravels above. 
It is only the middle member that is fossil-bearing. The sediments evidently 
were all water-borne, as the cross-bedding indicates, and the stream action 
must have been vigorous; this is also shown by the fossils. No associated 
skeletons were discovered, only skulls and scattered bones, many of them being 
water-worn. While the deposit is not prolific in fossils, nevertheless it yielded 
some good specimens which identified the formation as of Lower Oligocene age. 


On our first day the afternoon's prospecting netted quick results. Every- 
one took a different section. I had a pocket full of turtle shell, which was 
abundant, but as I worked around a comer I saw Shackelford on his knees 
scratching at the earth with his pick. He said that he had found some large 
bones, which I identified as rhinoceros. Berkey joined us, and together we 
traced the bed of a stream which millions of years ago had nm upon the stir- 
face. We were looking at a cross section of it and could see the successive 
layers of heavy gravel, small pebbles, sand and fine silt. It was easy to follow 
the course, and near the spot where Shackelford had fovmd the rhino bones 
we could see that there was an abrupt drop. Below was a heterogeneous mass 
of pebbles and large stones. This indicated a pool at the base of a small water- 
fall or rapid. An animal that died in the upper reaches of the stream very 
probably would be carried into the pool, sink to the bottom, and be covered 
with silt. Therefore, Berkey suggested that we dig into the bank at this point. 
In less than five minutes I had located a jaw and, directly below it, a large 
skull. Then Granger appeared upon the scene and put an abrupt stop to my 
excavations. Meanwhile Shackelford had discovered a beautiful rhinoceros 
jaw, partly embedded in the surface but in plain sight, and a set of teeth of a 
small artiodactyl. 

Instead of spending one day at this great bluff we remained three, because 
every time that Granger started to remove a skioll he discovered another a few 
inches from it. The ancient pool proved to be a veritable mine of fossils, but 
we simply had to get away if we were not to expose oiu-selves to serious danger 
from early snowstorms. It was difficult to leave a beautiful skull which 
actually could be seen, but it might take weeks to exhaust the possibilities of 
the deposit. 


Granger spent almost every daylight hoxir in the "Hole," as we called 
this particular spot on the side of the bluff, even having his tiffin sent up to 
him. At first he allowed human visitors, but one after another we did some- 
thing that inciured his palaeontological displeasure, and were ordered off the 
premises. Finally only our camp dog, Mushka, and the two pet choughs 
remained to keep him company. On the second day Mushka tipped over a 
tray of bones and was banished. The choughs behaved themselves fairly well 
and amused him greatly, for they were continually getting their glossy feathers 
so covered with flour paste that they covild hardly fly. But at last one of the 
birds committed an unpardonable sin. Granger removed an aknost perfect 
skull which had only a tiny piece of bone gone from one side. After nearly an 
hour's search he discovered the missing fragment and carefully pasted it in 
position. The instant his back was turned one of the crows hopped on the 
specimen, picked off the bone and swallowed it. Granger never forgave the 
bird, and even after he returned to Peking he was still gnimbling about it as 
he packed the skull for shipment to New York. 

I spent a good deal of time searching for fossils along the sides of the 
bluff, which extends many miles to the west. It was fascinating to wander up 
and down this "cross section of the ages." One could trace the courses of many 
stream sediments, almost aU of them productive of fossils. Turtle shells were 
abundant, and we found remains of a huge tortoise, Testudo insolitus. 


The fossil fauna of the Ardyn Obo formation which we obtained in our 
three days' stay was limited but interesting. Most important were the skulls 
of the aquatic "rhinoceros," Cadurcotherium ardynense, the extreme evolution- 
ary stage in the family Amynodontidae. A small perissodactyl representing 
a new genus and species, Ardynia prcecox, was represented by teeth and jaw 
fragments, and an interesting carnivore, Cynodictis, was discovered; also two 
genera of Artiodactyla, Eumeryx and Schizotherium, were represented by frag- 
ments. In 1923 we spent a week at this locality, and a more complete collec- 
tion was made, so that the relationship of the fauna could be determined. 
Doctor Matthew believed that it is somewhat older than the Hsanda Gol 
formation and remarked, 1923, concerning the fauna that "Its nearest geo- 
graphic affinities . . . are with western Europe, not with the United States."^ 


On a rocky ledge just under the rim of the bluff I discovered an eagle's 

' Matthew, W. D., and Granger, Walter. 1923. "The Fauna of the Ardyn Obo Formation." Amer. 
Mus. Novitates, No. 98, pp. 1-5; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., 
I, No. 17, pp. 1-5, 1926. 


nest — a great pile of sticks and branches which must have been carried a long 
distance, since we had seen no wood for miles. Near the nest were the remains 
of at least twenty antelopes, some of them being full-grown bucks, and the 
bones of kangaroo rats and other small desert animals were scattered over the 
hUlside. A pair of duck-hawks which lived high up among the rocks probably 
were responsible for much of the damage to the smaller mammals. 


I was extremely anxious to leave Mongolia, because of the danger of 
early winter storms. A heavy snow, which might come any day, woiild obscure 
the trail and keep us in camp until it melted. Then the terrain would be 
sticky mud in the depressions, and very diificiolt for the cars to negotiate. 
During the winter, cars can run without great difficulty, because the ground is 
solidly frozen, and the dry snow is swept from much of the desert, lying only in 
patches and drifts, but the late spring and early winter storms make motor 
travel well-nigh impossible, for the ground is warm and soft and the snow melts 
after a few days; usually before the terrain has dried another blizzard arrives. 

The other members of the Expedition were at a loss to understand my 
anxiety because they did not realize the sudden changes of Mongolian weather. 
On September 1 2 the day was so warm that it was uncomfortable to sit in the 
tents, and at night we slept under light blankets ; the thermometer registered 
78° F. at midnight. Shortly after daylight it began to rain and the temperature 
dropped thirty degrees in a few hours. The Expedition donned its ftir clothes, 
and the men began to realize that I had not been wrong in predicting the early 
advent of winter. Cold and bitter wind followed us back to Kalgan. 



We broke camp at Ardyn Obo on September 13, 1922, shortly after noon. 
I had waited as long as possible, but even then Granger was stiU working on 
his specimens and carried the latest skull in his lap for the remainder of the 
day; he would not trust it to a box in the car until the paste bandages had 
dried sufficiently to form a firm, hard shell. After a few miles the trail climbed 
out of the Ardyn Obo basin and led up and dowTi bare, hard-rock hUls, which 
looked like the solidified billows of an ocean in a mighty storm. About twenty- 
five miles from Ardyn Obo we reached a very large Chinese temple, Gatun 
Bologai, which was occupied by Mongol lamas. It has no element of Mongol 
architecture, and up to 191 1, when the Chinese were expelled from Mongolia, 
it probably had only Chinese inmates. The following year, 1923, we made use 
of the temple as a depository for food and gasoline. We pitched our camp 
September 13, 1922, ten miles southeast of the temple in a dry stream-bed 
where there were several good-sized elm trees. That night we had a blazing 
log fire, the first since leaving the forests of Sain Noin Khan. 


Various kinds of bad going were encoimtered during the next morning, 
but in the afternoon we sped at thirty miles an hour over a great rolling plain. 
At the western end of it the geologists were thrown into a fever of excitement 
by the discovery of a great deposit of Permian limestone, richly fossiliferous. 
These beds formed a part of Dr. A. W. Grabau's predicted Mongolian geosyn- 
cline or "earth trough" which ran through central Asia eastward to the Pacific 
at Vladivostok. During at least two long geological periods it had been filled 
with marine waters. Concerning this discovery, Doctor Grabau writes: "It 
may be mentioned in passing that the Mongolian geosyncUne in Dinantian 
and Permian times was outlined upon the maps purely because of the require- 
ments of palaeogeography, and without positive knowledge of the existence of 


marine strata of this age in the region included in the geosyncline. Subsequent 
discovery, by the Third Asiatic Expedition, of both Dinantian and Permian 
strata with rich marine faunas in the very heart of the region thus outUned, 
demonstrated the soundness of the principles on which the construction of 
these maps is based." 

The formation was designated Jisu Honguer, after the Mongol name for 
the district. The geologists could spend only two hoiirs at this spot during the 
first trip, but they visited it again both in 1923 and 1925, making a large col- 
lection of the invertebrate fossils. The formation occupies a narrow strip bor- 
dered by fatdt contacts between the granite hills on the south and the gray- 
wacke series on the north. Farther eastward, the entire series passes under 
sediments which carry a rich fauna of later Eocene vertebrates. 


Eighteen miles to the eastward from Jisu Honguer, Granger and I, who 
were in the leading car, saw an escarpment where characteristic Tertiary beds 
were exposed. Descending to the lowland we came to an area in which deposits 
of red, greenish, and purple clays were prominent. The nearest exposure is 
only a few yards from the road, and, after directing the other cars to proceed 
slowly along the trail, we walked over to the low ridge. Both of us immediately 
found fossils in great abundance. Bones of rhinoceros type were spread over 
the ground and we saw at once that it was one of the most remarkable fossil 
deposits that we had discovered in all Mongolia. We selected some of the best 
specimens from which to identify the locality and were on oiu* way back to 
the car when I saw a long bone partially buried in the earth. I called to 
Granger and in five minutes it was evident that here was a complete titano- 
there jaw with all the teeth in position. A real problem presented itself. 
Should we leave it, or spend the day or two that it wovild require to paste and 
bandage it properly? At any moment a snowstorm might tie up the expedition 
for days and I decided that it was unwise to stop. 

Granger said that he could remove one complete tooth row which wotild 
serve for positive identification of the specimen and for comparison with the 
American titanotheres. It would fall apart in many pieces, no doubt, but he 
could fit them together at the laboratory in Peking. The next half -hour saw 
an example of heroic methods in fossil dentistry. Every fiber of Granger's 
collector's soul rebelled against the crime he was committing upon a priceless 
specimen, and his groans as each tooth was extracted indicated as great pain 
as the titanothere itself wovild have experienced had it been alive. When the 
thing was done we carefully covered the remainder of the jaw, took bearings 
upon its location, and went on to join the men who were impatiently awaiting 
us. Farther along the trail we passed several other exposures, which evidently 


were a part of the same formation and indicated a vast region for future investi- 
gations. Granger was convinced, and correctly so, that this was a western 
extension of the Irdin Manha basin, in which we had fotmd the first fossils on 
the Kalgan-Urga road. The formation was formally designated the Shara 
Murun, and proved to be of Eocene age, but slightly younger than the Irdin 
Manha. The red-banded beds at the base probably are the equivalent of the 
Irdin Manha. 


Beyond the exposure the trail climbed to the top of the eastern edge of the 
basin. A short distance from the escarpment, in a small hollow, is a well con- 
taining clear, delicious water. It is known to the Mongols as Ula Usu, the 
"Well of the Mountain Waters." 

For eleven miles eastward we drove over the level Gobi erosion plane to 
the opposite side of the broad mesa, where a prominent escarpment exposed a 
few hundred feet of white and reddish sands and clays. Not far from the trail 
stood a small temple. Baron Sog-in-Sumu. Although we stopped for only a few 
minutes, we found the priests most cordial in their welcome. In future years 
this temple became a valuable storage depot for our gasoline and specimens. 


From the temple we could look eastward across the broad valley of the 
Shara Murun River. The stream was shallow and easily crossed in 1922, but in 
1925 it was swollen by heavy fall rains and became a formidable obstacle to the 
motor cars. The valley is about three hundred feet deep. Our direction on 
the eastern side changed to S. 80" E., the trail taking us over another great 
erosion plain, more rolling than that of Shara Murun. 


On the morning of September 15, we had a foretaste of what I had been 
expecting every day. Rain and bitter wind sent us into our heaviest clothes, 
but it was impossible to keep either dry or warm. Hour by hour the weather 
grew colder and the rain changed first to sleet, then to snow. We lost our way 
a dozen times and finally had to draw up in the shelter of a bank in a dry river- 
bed. The snow came so thickly that it was impossible to see. I had about 
decided to camp when the storm abruptly ceased. The ground was still warm, 
so the snow soon melted and again it was possible to find the tiny trail. We 
camped near half a dozen yurts under the lee of a great rock spine over which 
the wind howled like a charge of Mongolian demons. The entire Expedition 
ate dinner wearing fiu* gloves, and we immediately sought warmth in our 








• 5.Vi> 


Osllih Basin. 1923. 


Ansa Bogdo. 1923. 


Holding the newly found duplicate tooth of the 
coryphodont, Eudinoceras mongoliensis. Irdin 
Manha, 1923. 


Fi"e Antelope Camp, 1922. 



During the next day's travel, the country gradually changed into the 
typical southern Mongolian grasslands and we came to the outlying fields cul- 
tivated by Chinese farmers. They showed at first as isolated patches with 
strips of vmbroken ground between, but a few miles farther south the fields 
appeared as a continuous line of standing or half-harvested grain. Except for 
the blue-clad Chinese peasants and the occasional mud-walled houses, we 
might have been looking across the grain fields of North Dakota. 

In a single day we had come from the arid reaches of a semi-desert into the 
midst of a prosperous farming community. On the very outskirts of cultiva- 
tion we saw a few Chahar Mongols, but as the fields increased these disappeared 
and none but Chinese remained. Just as the white men in America drove the 
Indians farther west, so the Chinese are pushing the Mongols out of all the 
southern grasslands of Mongolia. The Mongols are essentially a pastoral 
people. By nature they are not agricviltural, and centuries of tradition have 
made them nomads. They seem not to be able to change their life even as they 
see themselves driven out of their finest grazing lands by the steady encroach- 
ment of the Chinese farmers. Year by year areas of new land are opened for 
settlement by the Chinese. On the Kalgan-Urga trail, the most distant culti- 
vated locality in 19 18 was seventy miles northwest of Kalgan; in 1930 it had 
been advanced to one himdred and fifteen miles. Soon all of the Inner Mon- 
golian grasslands will have been occupied and the northward extension of cul- 
tivation brought to a halt at the edge of the desert. Although irrigation 
undoubtedly could be carried on to advantage in many parts of the Gobi, I 
believe that it will be many years before this is attempted, unless the much- 
talked-of railroad to Urga materializes. The soil of the grasslands appears to 
be deep and rich. There is an abundance of rainfall, and ponds are frequent. 
Millet, wheat, oats, barley, flax and potatoes are the principal crops. 

The last himdred miles through the area of cultivation were difficult. 
Nowhere in our entire exploration in the Gobi had we encoiuitered such diffi- 
cult traveling. More damage was done to the motors during the first day up 
the Wan Ch'uan Pass, and in the last two days of owr return trip, than in all the 
other three thousand miles. Long trains of Chinese carts pass on their way to 
Kalgan, and their spike-studded wheels work havoc with any road. There 
were mud and deep ruts in the valleys and rocks and ruts on the hills. 

Late in the afternoon of September 18, we reached Miao T'an, the Chinese 
village thirty-four miles north of Kalgan, which we had left just five months 
before. It was a great day for us. We had a dinner of Chinese food at the inn 
and from the proprietor learned much of the political news of China. We were 
dirty and travel-worn but happy, for we knew that the Expedition had been 
an unqualified success. 



I engaged a cart to take the heavy baggage to Kalgan so that our cars 
might go down the Pass as light as possible. It was well that we had no excess 
weight, for I have never known the Pass to be in worse condition. It was a 
nightmare for motors and ovir average speed was only four miles an hour. We 
had made so close an estimate on our gasoline supply that all the cars reached 
Anderson, Meyer and Company's compound except Colgate's. A can of 
gasoline had to be sent to bring that one car the last half mile, from its stopping 
place just iBside the city gate. 

Larsen was in Kalgan and we had much to tell him. His first question 
was: "Did you get the wild ass you were chasing when we left you on the crest 
of the hill?" The first thing we asked him was how soon we could get a bath. 
Berkey and Morris stayed with Larsen; the rest of us went to the British- 
American Tobacco Company's mess, where the doors are always open to 
travelers from the interior. 

Our efforts to meet the requirements of civilization were almost pathetic. 
Each one of us had some article of adornment which he had been cherishing 
for the home-coming. Shackelford appeared in a wonderful blue shirt. I had 
a purple necktie, while Granger and Colgate each produced a new pair of 
shoes. When we came into the dining-room for tea, where half a dozen 
visitors had assembled to welcome us, we all felt decidedly vmcomfortable. 


We left Kalgan for Peking on September 21, exactly five months from the 
day we started for the great plateau to test a theory. The day following our 
arrival at Peking, the Expedition staff gathered in the great drawing-room of 
oiir headquarters to compose a cable to the American Museimi of Natm^al His- 
tory, which would announce the principal results of the summer's work. I 
suppose no moment in an explorer's life is more satisfying than that when he 
makes public for the first time the story of the months during which he has 
been lost to the civilized world — if that long isolation has produced results 
which justify the expense and human effort ! 

In perspective 

Another interesting experience is the gradual gaining of a new perspective 
on what he has done. Although none of our staff ever lost his enthusiasm for 
the work, still the very closeness to it, and the continued discovery of new 
facts of vast scientific importance, at times tended to make it almost common- 
place; the daily fatigue of overworked muscles and nerves, the discomfort and 
frequent hardships, sometimes made one wonder if it was all worth while. But 
in a comfortable home, with rested brains and bodies, with the work done and 


the strain and anxiety ended, we coiald begin to understand, as a whole, the 
significance of what had been accomplished during the summer. 

Roughly summarizing the results, we realized that the work had resulted 
in an almost overwhelming amount of new information in geology, palaeontol- 
ogy and geography. In spite of the pessimistic predictions before our start, we 
had opened a new world to science; in the rocks of Mongolia we had before us 
a splendid record of earth history and an equally wonderful record of past 
life. Our new geological formations fitted admirably into the general geo- 
logical column of the world. The understanding of vertebrate evolution and 
dispersal would be immeasurably widened by our palaeontological discoveries. 

The traverse 

We had traveled from the seacoast straight across Mongolia to the forests 
of the northern edge, up to the Arctic Divide, and could visualize the country 
as though it were a relief map. Coming from the seacoast, the low sand and 
silt plains stretch right up to the base of the mountains through which the 
Nan K'ou Pass gives access to a new level nearly two thousand feet higher. 
Crossing this smaller and higher plain we had come to a new barrier, the Pacific 
Divide, topped by the outermost rampart of the Great Wall; a barrier five 
thousand feet above the level of the sea. This is part of the southern edge 
of the Mongolian plateau, which is one of the great interior basins of central 

A thousand miles northward across the plateau stands the Arctic Divide 
separating the drainage into Siberia from the waters that flow south into the 
desert. Between the Pacific and Arctic divides lies a great cradle-like sag, 
descending gradually to the depression at Iren Dabasu, where the elevation is 
not more than three thousand feet above sea-level. From all sides the land 
slopes to the interior. This large basin-like region has not a simple floor, but 
it is a complex of smaller basins, each one of which is smooth and flat, and 
separated from contiguous basins by roUing hills or semi-mountainous ridges. 
It is these open stretches of level groimd in the secondary basins that are 
called "gobi" by the Mongols. 

We had found that wind, and to some extent rain-wash erosion, had swept 
the lighter material from the surface, and that the Gobi is essentially a "rock 
desert," with sand-dune belts only in restricted areas. Day after day we had 
traveled over the nearly barren stirface of rock. 

The geology 

The geologists had learned that the geologic formations of Mongolia 
consist of two grand divisions: one an exceedingly complex series of ancient 
rocks carrying the story back to the very dawn of geologic history; the other 


a simple series of sediments recording the last chapter. Between this ancient 
series and the later one, a long interval is lost. 

We had discovered a gigantic granite bathylith, possibly the most exten- 
sive in the world, vmderlying the later sediments and frequently coming to the 
siirface. We had learned that since Jurassic times Mongolia has been con- 
tinuously dry land, with no period of submergence beneath the sea; that 
across the plateau had extended a great geosyncline as predicted by Grabau; 
that there was no evidence of glaciation other than that of a strictly alpine 
type which had not reached the plains; that there is reason to believe that 
climatic cycles or fluctuations have occurred in Mongolia from late pre-Cam- 
brian up to the present time. The geological column of Mongolia had been 
partially filled in by the identification of many geological systems. Due to 
the indefatigable efforts of Professors Berkey and Morris, we had obtained 
the longest detailed topographic route map and continuous geologic sections 
ever made on reconnaissance. From the standpoint of geologic science alone, 
the Expedition had been a conspicuous success. 

The palaeontology 

Pateontologically, our results were quite as significant. In a vast region, 
the life history of which was absolutely unknown, we had discovered some of the 
richest fossil fields of the earth. In this first season's work twelve vertebrate 
fossil-bearing formations had been identified and named, beginning with the 
oldest, as follows: 

Cretaceous Period: The Ondai Sair, Oshih (Ashile), Iren Dabasu, and 
the Wan Ch'uan formations. 

Upper (?) Cretaceous Period: The Djadochta formation. 

Eocene Period: The Irdin Manha and Shara Murun formations. 

Oligocene Period: The Ardyn Obo, Houldjin, and Hsanda Gol forma- 

Miocene Period: The Loh formation. 

Pliocene Period: The Hung Kureh formation. 

Nearly two thousand specimens had been obtained from these forma- 
tions, representing several new families and subfamilies and a great many new 
genera and species. These indicated that the Mongolian fossil fauna would 
fall into three categories: First, animals that originated in Asia and never 
strayed outside its limits; second, animals that originated in Asia and migrated 
to America or Europe ; third, animals that originated in America and migrated 
to Asia. 

The zoology 

The zoological collections niimbered several thousand mammals, which, 


together with those that I had obtained in 19 19, gave us an unrivaled series 
from north and central Mongolia. Subsequent study showed that most of the 
desert and grassland species had a very wide distribution, as might be expected 
in a region where there are few natural barriers. The ranges of some forms 
were extended for many hundreds of miles beyond previous records. 

The photography 

Mr. Shackelford had obtained twenty thousand feet of film and many 
stiU photographs, showing every phase of the Expedition's work, as well as a 
complete story of Mongol life. Such a permanent record of a vanishing cultiire 
is of enormous value. 

The motor transport 

The fact that we had been able to use motors successfully in this remote 
region immediately brought commercial results. We had hardly returned 
before fur traders and wool buyers arrived to ask how they could reach out- 
lying stations with automobiles. In a short time it became evident that our 
Expedition had opened central and western Mongolia to motor transport. As 
a matter of fact, within three years a hundred cars were running where not one 
had been before. 


At the staff meeting we projected a complete program for the following 
year's work. It was evident that we must concentrate on palaeontology in 
order to take advantage of the great fossil fields which had been discovered. 
This could be done only by asking the Museum to send us three additional 
expert fossil collectors. Doctor Berkey found it necessary to return to New 
York at once, to take up his work at Columbia University, but Professor Morris 
was to remain in Peking and spend the winter writing, and studying Russian. 
On the next Expedition he would devote himself to special areas where the 
geology needed further study. 

In November, Doctor Granger would proceed to Wanhsien, Szechwan, to 
complete the collections which he had made there during the winter of 1921- 
1922. I would remain in Peking to make the necessary preparations for the 
coming stimmer's exploration. Mr. Shackelford would return to New York 
to arrange and edit the twenty thousand feet of motion-picture film. Mr. 
Colgate was to continue his trip around the world, of which the expedition to 
Mongolia had been a part. 

In addition to the three extra palasontological collectors, it would be 
necessary for the next Expedition to include two motor experts to handle the 
transportation. Since we would cover virtually the same region and make the 


next summer one of intensive work, it was deemed advisable not to take Mr. 
Shackelford, the photographer. 

We discussed most carefully the relative palasontological importance of 
the various fossil fields, and made a paper plan of the amoimt of time which 
could be profitably devoted to each. Our estimates were so accurate that with 
one exception we were able to follow the schedule aknost to a week. 

All our equipment, including the motors, was brought back to Peking 
for renovation. Each car was completely "taken down" and overhauled. We 
were amazed to find that even after the grueling travel most of the car-parts 
showed little wear, and in only one or two instances were replacements neces- 
sary. It was a wonderful record for the Dodge Brothers automobiles. 


On October lo, I received word that the caravan had reached Chap Ser, 
but was afraid to proceed to Kalgan because of the brigands. Bayard Colgate 
and his cousin, Gilbert, who had joined him in Peking, volunteered to escort 
them through the dangerous territory. This they did without incident. The 
caravan had arrived ten days before the stipulated date, and Merin and the 
others were given a month's pay as bonus for their excellent work. I pur- 
chased a new yurt, flour, millet and other necessities, and sent them back with 
the camels to spend the winter near Chap Ser, a hundred miles north of Kalgan. 
The animals were thin and tired after their summer's travel, and Mongolia 
offers but poor feed during the winter; nevertheless, it appeared best to keep the 
camels, as it was impossible to sell them to advantage. The winter's transport 
was just beginning, and all caravans wished to start with fat, imtired camels. 

The collections were repacked in Peking and shipped to New York by 
water. They reached the Museimi without the loss of a single box, and with 
every specimen in as good condition as when it had been removed from the 
rock in the Gobi Desert, thirteen thousand miles away. 


On his journey up the Yangtze River to Wanhsien, Granger had the usual 
experience of being continually fired on by brigands and soldiers. His whole 
winter was prevented from being monotonous by the inter-provincial war 
which is always in progress on the upper river. He did much more than 
palagontological work, for his presence in the village of Yen-ching-kou calmed 
and assiu-ed the inhabitants to such an extent that few of them left their homes, 
even when others fled to the hills. Yen-ching-kou is only twenty miles from 
Wanhsien, a city of considerable importance. Granger camped in a temple 
and carried on his work by buying specimens from the natives. He writes of 
the peculiar conditions of the deposit as follows: 



"The fossils at Yen-ching-kou occior in pits, distributed along a great 
limestone ridge some thirty or forty miles in length and rising above our camp 
over two hundred feet. These pits are the result of the dissolving action of 
water on limestone, and some of them have a depth of one hundred feet or 
more. They are of varying sizes, averaging say six feet in diameter, and are 
filled with a yellowish and reddish mud, which is, I take it, disintegrated 
limestone. The fossils are found embedded in the mud at varying depths, 
usually below twenty feet. A crude windlass is rigged up over the pit, and 
the mud dug out and hauled to the stirface in scoop-shaped baskets. At 
fifty feet it is dark in the pit, and the work is done by the light of a tiny oO- 
wick. It is fossil-collecting under the most adverse conditions imaginable. 

"The excavation of the fossils has been going on for a long time — possibly 
some generations — and it is a considerable business. Digging is done only 
in the winter months. 

"One has to be let down with a rope around his waist and with two or 
three men at the windlass. The natives climb up and down the rope hand 
over hand, but it requires practice and agility to do this. You'd be shy one 
palaeontologist if I tried it!"^ 

Until Dr. J. G. Andersson began his splendid work with the Geological 
Survey of China, knowledge of the palaeontology of China rested almost 
entirely upon the work of Schlosser.^ All Schlosser's material was purchased 
in drug-shops and consisted of teeth and fragments of bones, without locality 
information. Thus, when Andersson began to discover -fossils in situ he had 
a virtually untouched field before him. From his work and that of Schlosser, 
there is evidence of at least two distinct faunas in north China, probably 
separated by the Tsinghng Mountains of Shensi. 

To the north is the so-called Hipparion fauna of Pliocene age, characterized 
by an abundance of horses. To the south is what we have named the Stegodon 
fauna, for the remains of these primitive elephants are common. 

During the three winters which Doctor Granger spent at Yen-ching-kou, 
he made a splendid collection, which was studied by Dr. W. D. Matthew, 
and comprises genera of the following orders: Primates, 2, Bunopithectis and 
Rhinopithecus; Camivora, 7 ; Rodents, 2 ; Proboscidea, i ; Perissodactyla, 3 ; 
Artiodactyla, 7. 

Regarding the affinities of the favina Doctor Matthew, 1923, remarked: 

"The above list is remarkable, as a cave or fissure fatma, for the scarcity 
of rodents (other than Rhizomys) and small camivora. While the remains 

'Andrews, Roy Chapman. 1922. "Politics and Paleontology," Asia Magazine, XXII, No. 5, May, 
P- 363- 

* Schlosser, Max. 1903. "Die fossilen Saugethiere Chinas nebst einer Odontographie der recenten 
Antilopen," Ahh. Math.-Physik. Kl. k. hayer, Akad. Wiss., XXII, pp. 1-221. 


of large animals are abundant and varied, the bamboo-rat is the only rodent, 
except for a single hare jaw, and no small mustelids or viverrids appear. It 
is no less remarkable that no trace of Equidae is found in it, nor of camels, 
giraffes, typical Canidai or machaerodonts. This, coupled with the abundance 
of tapirs and deer, may point to a heavily forested condition. The abvmdance 
of Stegodon, the entire absence of Elephas, and the presence of Chalicotherium 
are the only observed indications of Pliocene age ; for the most part the fauna 
appears to be quite closely related to modem species and might well be con- 
sidered Pleistocene. The faimal affinities appear to be principally Chinese, 
partly Malayan, not much Indian; there is nothing especially suggestive of 
North American or of Siberian affinity. A more carefvil comparison and 
identification of the whole fauna, especially of the smaller ruminants, might 
show a clearer differentiation from the modem species than we have observed 
in this preliminary study, but could hardly alter materially the geographic 
and environmental affinities of the fauna. It is such a fauna as one might 
expect to find in the valleys of southwestern China at any time before the 
appearance of civilized man, and imder climatic conditions similar to those 
now prevalent. The effect of the clearing and cultivation of the valleys and 
the lower slopes of the hills by man has been, broadly speaking, to drive the 
smaller animals to the mountains and to exterminate the larger ones. Some 
of the extinct types have left relatives, more or less distant, in the jungles of 
southeastern Asia, more resistant to human encroachment than the Chinese 
hills. But the tapir, rhinoceros, gaur and Stegodon of the Yen-ehing-kou 
fauna, although their nearest existing relatives are of tropical habitat, do not 
necessarily indicate a warmer Pliocene climate in China. They may quite 
well have been species adapted to a temperate climate, such as is more defi- 
nitely indicated by the geographic affinities of the remainder of the fauna."' 

• Matthew, W. D., and Granger, Walter. 1923. "New Fossil Mammals from the Pliocene of Sze- 
Chuan, China." Bull. Amer. Mus. Nat. Hist., XLVIII, Art. xvii, pp. 397-398; Preliminary Reports, Central 
Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 15, pp. 563-598, 1926. 



My urgent request to the President of the American Museum for three 
additional expert fossil collectors to accompany us on the 1923 Expedition met 
with an immediate response. Messrs. George Olsen, Peter Kaisen and Albert 
Johnson were selected. The first two had been on the Museum's field and 
laboratory staff for many years ; the last had worked with Mr. Bamum Brown, 
collecting dinosaiu-s for several seasons in Alberta, Canada. 

To handle the motors, I was forttmate in obtaining the services of Mr. J. 
McKenzie Young, of the U. S. Marine Corps, who was appointed Chief of 
Motor Transport ; he had Mr. C. Vance Johnson for assistant. Both men were 
expert motor engineers and I knew that this important branch of the Expedition 
would be in competent hands. 

The native staff remained the same as that of the preceding year, except 
that otir mess "boy," Kan Chuen-pao, or "Buckshot," had been promoted to 
be a palagontological assistant to Doctor Granger. He had demonstrated such 
ability and interest in handling fossUs at Yen-ching-kou, during the winter, 
that Granger realized we were losing real talent in keeping him as a mess boy. 
His judgment was amply confirmed, for "Buckshot" became one of the most 
valuable men we have ever had. Another yotmg Chinese, Liu Hsi-ku, was 
added to the staff as a mechanical assistant. Later, he, too, graduated into 
fossil work, and with "Buckshot" was sent to the American Museum at the 
end of the summer for a year's laboratory training. 


As ovu first investigation was to be in the Iren Dabasu basin, two hundred 
and sixty miles from Kalgan, I did not send the caravan away until April 2. 
The staff left Peking April 17, 1923, the date of our departure the former year. 
I hoped to strike the fortnight of fine weather which usually marks the begin- 
ning of the Mongolian spring, and to get well at work before the return of cold 



weather about the seventh of Alay. We were a few days too earl}', however, 
and found the siunmits of the hills about Kalgan white with new-fallen snow. 
Mr. Larsen arrived from Urga late in the same afternoon and reported that 
his motor had crossed the last hundred miles on the plateau with the greatest 
difficulty, because of the soft snow which formed a mass of gluelike mud. 

I decided to wait a few days until the terrain improved. Moreover, Mr. 
Larsen intended to return to Urga immediately, with a car full of silver dollars, 
and he was very anxious to travel with us for the sake of safety. A week earlier 
two Russian motors had been stopped by Chinese soldier brigands, dressed in 
uniform. The passengers had lost a valuable cargo of furs and all their personal 
belongings, even being stripped of the clothes they wore. One man was killed. 
Several caravans also had been attacked by bandits during the past month. 
All the robberies had occurred in the district just north of Kalgan, which is 
under the control of the Tutimg of Chahar. His efforts at maintaining the 
safety of travelers in the area of Chinese cultivation were as futile as those 
of other officials throughout China. Every few weeks several brigands would 
be brought to Kalgan by soldiers, taken to the dry river-bed which runs through 
the center of the city, and shot. Sometimes they caught real bandits but 
just as often innocent peasants were executed. The "trial" was an empty 

On April 19, two days after our arrival in Kalgan, we left for the great 
plateau. Our passports were inspected at the barrier where Mr. Charles Colt- 
man had been murdered by Chinese soldiers during the winter. Just a year 
before he had accompanied us to Urga, and a wave of sadness swept over me 
as I realized that the friend with whom I had lighted so many camp-fires on 
the Mongolian plains never would travel those trails again. At the summit of 
the Pass we halted to look back over the vast complex of shimmering badlands 
through which we had made our way. There, we bade good-bye to China 
and turned to the far-flung reaches of Mongolia beyond the wall. Mongolia, 
with the romance of a glorious tradition through which move the shadowy 
forms of Ghengis Khan and his world-conquering barbarians! MongoUa, a 
land of painted deserts dancing in mirage ; of limitless grassy plains and name- 
less snow-capped peaks; of vmtracked forests and roaring streams! Mongolia, 
a land of mystery', of paradox and promise ! 

The first night we spent at the little Chinese inn of Miao T'an, thirty- 
four miles from Kalgan, for I had no wish to camp in the brigand-infested 
hills. There was a company of soldiers quartered in the village, and when 
we were about to start in the morning, their commander sent an officer to say 
that for our "safety" he had despatched a dozen men beyond us on the road. 
He ended by remarking naively, "Please don't shoot my soldiers!" I assured 
him that we would be careful whom we killed, but our guard were taking no 




^ - ^. - 





-c »^*" 

'^^■^ ^ 


.* ^ 

AKi>A lUK.lili, ^Hl.^Ul^"l.. . 



chances. When we passed them about five miles beyond the inn they sounded 
a bugle and drew up beside the road with a flag prominently displayed for 
purposes of identification. As a matter of fact, we did see some twenty 
mounted brigands before we were out of the cultivated area. They lined the 
hills on both sides of the road, but our cars, bristling with rifles, did not impress 
them favorably and they left us severely alone. 

Those of us who had been in Mongolia before were astonished at the 
extension of cultivation dimng the last year. A new tract had been opened 
by the Chinese Government, and tilled lands now reached to Chap Ser, one 
himdred miles from Kalgan. At Hallong Ossu, near Chap Ser, there was a 
Swedish mission station in charge of Rev. Joel Eriksson. In 1927 it had to 
remove farther into the grasslands, for the Mongols had all been driven out 
by Chinese encroachment. 


I never have seen the Kalgan-Urga road as fine as it was that spring. 
Because of the previous two years of war, cart traffic had ceased and the ruts 
had all been worn down by weathering and the padded feet of camels. When 
we crossed the previous year, the plains were deserted. Monasteries lay in 
ruins; yurts were absent, and no flocks of sheep or goats drifted up the sheltered 
valleys. If a solitary Mongol sighted the cars, he fled like a frightened animal. 
With the kaleidoscopic changes which come to these countries of the Orient, 
the land had assumed the normal aspects of peace in one short year. Yurts 
clustered near almost every well, like giant beehives; herds fed quietly on the 
grasslands, and picturesque caravans of majestic camels passed us on the road, 
carrying loads of skins to Kalgan or merchandise northward to Urga, the city 
of the Living Buddha. Winter furs were coming down from Urga on every 
automobile, and crowds of Chinese merchants were going up. Ten to fourteen 
men, and as many bedrolls, were the usual load for a five-passenger Dodge 
Brothers car. Each one paid "Mex. $50," the equivalent of fifty Mexican 
dollars, then about $25 gold, for a chance to ride, and it was his affair to find 
a seat. As many as possible squeezed into the tonneau, and the less fortunate 
ones half lay on the bedding, which was piled on the running boards level 
with the top of the doors, their legs sticking out in all directions. At a little 
distance an auto thus loaded appeared like an enormous spider. The springs 
of the cars were blocked with wood and rubber, yet it is a miracle how they 
stood the test. 

We reached the first telegraph station at P'ang Kiang before dark. It 
had been a rtin of one hundred and forty-three miles from Miao T'an, the 
longest day's trip the whole fleet had ever made. I shot two antelope for meat, 
and our camp that night was a joyous one. Near the telegraph station a 


large mud-walled inn had been newly erected, and preparations were being 
made for heavy motor traflfic during the summer. 


We had confidently expected to find our caravan awaiting us at Iren 
Dabasu, but it was not there. I knew that Merin had taken a trail east of the 
main road, because it was safer and offered better grazing for the camels, yet 
Mongols who had traveled by the same way reported no sign of our caravan. 
It was easily identifiable, for the boxes were peculiar in shape and the big 
lead camel always bore the American flag. I was greatly worried, for brigands 
were so numerous in that region that I feared the caravan had been captiired 
and driven off into the desert. While food, which the boxes contained, would 
be of little interest to bandits, the camels themselves could be disposed of 
easily enough, and the three thousand gallons of gasoline sold in small lots to 
motor stations along the main road. 

But canny old Merin had not been caught, after all, and one evening a 
car roared into camp with the word that our camels were twenty miles away 
at the Lung Ku Shan (Dragon Bone HUl), and would be in next day. The 
Mongols arrived gleeful as children to be safely in camp with us. Merin said 
that when he left Chap Ser he heard that there were bandits ahead of him 
watching the trail. Therefore, he slipped off into the desert, traveling at night 
from weU to well and camping diuing the day in sheltered hollows where he 
could not easily be seen. His weather-tanned face beamed as he told how he 
had played hide-and-seek with the brigands and yet had filled the stomachs of 
his camels with some of the best grazing they had had all winter. Two of the 
weakest camels had died on the way to our camp, but the others were in fairly 
good condition and only needed an abundance of food to make them fit. 


Although Iren Dabasu is a desolate area of sand-dunes and sagebrush, 
it offers the most delectable food from a camel's standpoint. Sagebrush, 
thorny bushes, or almost anything as long as it is sufficiently dry and salty, 
offers him the maximum of gastronomic entertainment. He is such an extraor- 
dinary example of adaptation to environment that I never tired watching the 
herd at rest or on the march. 

The Mongolian camel is able to pick enough nourishment from the well- 
nigh invisible vegetation on a gravel plain to keep him alive, his great, fiat foot- 
pads carry him over loose sand as easily as though he were on snowshoes, and 
his ability to go for long periods without water makes him an invaluable asset 
to the desert nomads. Wherever possible the Mongols watered our camels 
every two days, but once or twice they had to go five days without a drink. 


But the camel is such a highly specialized beast that he is helpless away 
from his desert and cannot adapt himself to changed conditions. While his 
long legs and flat pads carry him over the sand at the steady pace of two and 
one half miles per hour, if he encounters mud or slippery grotmd he simply 
gives up and would die in his tracks without making an effort to save himself. 
His feet have no grip and slide in every direction ; then his long legs are a dis- 
tinct disadvantage, for they spread out, and once down it is difficult for him to 
rise. Rocky terrain cuts and tears his feet and he stands helplessly, uttering the 
most doleful wails. 

He is as timid as a mouse and as delicate as a child. He catches cold very 
easily, and if his long neck hair is cut or his winter's coat is pulled off too soon, 
a sudden change in temperature will put him on the sick list for days. The 
natives believe that when a camel sheds his hair, which he does at the end of 
June or in July, he is unusually weak. At that time he is virtually naked. If 
he is kept at work during the shedding process he will not grow as good a coat 
during the summer. In the Gobi it is difficult to find smtable packing material 
for fossils, but the camels solved this problem for us by furnishing their shed 

A camel's temper is uncertain, and as a rule he likes neither the odor nor 
the appearance of a white man. He will kick and bite viciously, but even worse 
is he when he manifests his displeasure by spitting. A green mass of half- 
digested evil-smelling vegetation is discharged in a spray. It requires days to 
remove the taint from one's body or clothes, and "once spit upon twice shy" 
was an adage which was brovight home to every member of the Expedition by 
sad experience. He can become accustomed to eating grain and dry straw, 
but his natural food is the vegetation of the desert. Unlike the single-humped 
dromedary, the Bactrian camel is a cold- weather beast, and work in the heat 
rapidly saps his strength. He will not graze at night, and for this reason, even 
in the winter, caravans usually rest during the day, start late in the afternoon, 
and travel until two or three o'clock in the morning. At night the animals 
are brought to the camp and are tied, kneeling, by their nose-strings, head to 
head on either side of a long rope. 

We loaded our camels only up to four hundred pounds. This is the 
maximtim weight that they can carry on a long trip, although for a short dis- 
tance five or even six hundred pounds is not too much. The strongest are the 
short-legged, heavy-bodied type. The long-legged beasts never seem to have 
as much endurance as the others. Those which are bred in a very sandy region 
have wider foot-pads, but these are also thinner and more easily cut by travel 
over rough groimd. 

We had only gelded camels in our caravan. The btills begin to rut in 
December and at that time are very dangerous. Only an experienced man can 


handle a rutting bull. Once a bull followed my motor for five miles. It hap- 
pened to be in heavy going where we could not make speed, and the enraged 
animal almost overtook us. Several times I thought it would be necessary 
to shoot him, for had he caught us he certainly would have done his best to 
annihilate the entire party. The period of gestation is thirteen months and 
calves are usually dropped in January. The cow then is unfit for work, and 
as she attends the calf for more than a year, she does not bear young oftener 
than once in three years. 

I always allowed the caravan leader to have for himself, as a perquisite 
of his position, such of the camel hair as we did not need for packing specimens. 
As the wool is worth about one hundred dollars (Mex.) a picid (133 lbs.), and 
as each camel sheds six or eight pounds, it was a lucrative job for a Mongol. 


At Iren Dabasu we camped on the same spot where our tents had stood the 
previous year. The desolate surroundings were even less attractive than usual, 
for the entire central Gobi was parched from ten well-nigh rainless months. 
Even the light showers, which in a few hours seem to completely change the 
face of the desert, had been absent during the auttmm, and there had been 
but little winter's snow. The salt marsh lay white and stark without a trace 
of water; the sand hillocks and the basin floor were vmtinged with green; it 
was a dreadful monotony of brown and yellow. 

The morning after our arrival, the three new men, Kaisen, Olsen and 
Johnson, set out eagerly with picks and collecting sacks to have their initial 
experience in the fossil fields of Mongolia. Morris, Granger, Young and I 
drove westward on a trip of exploration, to see if we could locate additional 
exposvu-es. Eight miles from camp we saw the familiar gray-white strata and 
stopped to prospect. Almost immediately teeth and fragments of bone were 
found scattered over the surface in a half-dozen places. Granger discovered a 
huge femur, half exposed by the action of wind and water and frost which were 
wearing away the rock particle by particle. 

As I walked over the ridge that day I had the same feelings which I sup- 
pose inspire every prospector for gold. It was a likely-looking place, and at 
any moment I might discover a discoloration in the rock or a tiny fragment 
of bone which would give the clue to a mine of palaeontological wealth. With- 
out a doubt there were himdreds of bones lying just beneath the surface. 
But where? If only my eyes could pierce that baffling surface and get a glimpse 
of what lay concealed! 

It is well-nigh hopeless to dig for fossils unless there are definite indica- 
tions of their presence. There must be some sign, a piece of bone "nmning 
in," or something upon which to fasten hope. Otherwise one might dig and dig 


A. SKULL OF Andi-L-icsarcliHS and of an American timber wolf. 

B. restoration of Andreii'sarchus under the direction of henry fairfield osborn, by mrs. e. r. fulda. 


Paint«l l)y Mrs. E. R. Fulda. 

Painted liy Charles R. Knight. 


and miss the greatest treasures by a few feet or even inches. Of course, one 
finds bones which have been broken by the action of weathering into a hun- 
dred fragments and which have no connection with other parts of the skeleton 
beneath the surface. These disappointments come most frequently in fossils 
deposited in an old stream-bed, where swift water has rolled and broken them 
before they could be buried in sediment and preserved. 

The Houldjin gravels, where remains of Baluchitherium are abundant, 
are such a deposit. Time after time I have felt my blood thrill with excitement 
at sight of a projecting bone, thinking that it might mean that a skeleton lay 
underneath, but after carefully working away the rock or earth only a useless 
fragment would be uncovered. My feelings about the Houldjin gravels, and 
all similar stream deposits, are a standing joke in the Expedition. 

I am hardly philosophical enough for a palaeontological collector; dis- 
appointments and successes send me too easily into the blackest depths or to the 
pinnacle of happiness, and particularly I cannot curb my impatience sufficiently 
when a specimen has been found. Walter Granger, or any of the other trained 
men, are content to work away grain by grain with a camel's hair brush, wait- 
ing for the specimen to develop as they go down. Theirs is admittedly the 
proper way to proceed, but pick and shovel methods, which at least give quick 
results, come more naturally to my restless spirit. Perhaps a complete skeleton 
or a priceless skull lies below that bit of projecting bone, and I simply cannot 
wait for days to know. Therefore, whenever one of the men is engaged upon 
the delicate operation of removing a specimen, the Chief Palaeontologist issues 
an ultimatum to the Leader of the Expedition, "Thou shalt not approach this 
sacred spot unless thy pick is left behind." 


In our brief siu"vey of the Cretaceous ridge west of Iren Dabasu, we saw 
enough to warrant sending over three men for a careful inspection. Each one 
of them immediately discovered important fossils, which in every case devel- 
oped into "quarries." Albert Johnson's proved to be the richest. His keen 
eyes were attracted by a fragment of bone not more than three inches long. 
By following this clew he gradually exposed such a great deposit that for a 
month he and Kaisen worked continually in a single spot. The fossils lay only 
a foot or two below the surface, but were so completely covered that except 
for the three-inch bit which gave away the secret, their presence would have 
been vinsuspected. 

In this deposit, bones of both flesh-eating and herbivorous dinosaurs, of 
several species and of many individuals, were piled one upon the other in a 
heterogeneous mass. They appeared to have been subjected to a swirling action 
when they were deposited, which led us to believe that this spot had been a 


backwater or eddy at the edge of the lake or stream. When the dinosaurs died, 
their bodies drifted into this bayou, where they came to rest, the flesh decom- 
posed, and the skeletons sank into the soft mud, eventually to be fossilized. 

On the shores of the ancient stream or lake grew a lush vegetation, for the 
dinosaurs' bones are of the bipedal, duck-bill iguanodont type which wallowed 
in the mud along the edges of watercourses where the plants were soft and suc- 
culent. Since this type was an herbivorous feeder and without means of 
defense, it must have been a prey of the carnivorous or flesh-eating dinosaurs 
which were contemporaneous. We found some bones of the carnivores mixed 
with the iguanodont remains. Very probably these fierce creatures, in their 
struggles with the duck-bill dinosaurs, had been drawn into the water and 
drowned, their bones becoming fossilized with those of their victims. Pre- 
sumably the lake or stream was surrounded by more or less open, savannah- 
like country, since we discovered small dinosaurs of the running, plains type 
like Ornithomimus. There were also crocodiles, turtles of the Trionyx type, and 
a few pelecypod shells. 

During the Age of Reptiles, with conditions particularly favorable to their 
development and with nothing except natural causes to check their increase, 
the extent of the reptilian life which swarmed in this region baffles imagina- 
tion. To-day, this nightmare world of the past is gone. In its place lie the 
silent, wind-swept ridges of the Gobi Desert, parched and blistering under the 
summer's sun, during the winter an area of arctic desolation. 

Not far from the quarry which Johnson opened, Kaisen and Olsen each 
found deposits where fossils were in abundance. Olsen discovered the com- 
plete hind limb of a large carnivorous dinosaur. The leg lay doubled up just 
as the great reptile had died millions of years ago. The remainder of the skele- 
ton had weathered out and broken into fragments, or, in those far-distant 
days, other flesh-eating dinosaurs may have torn the carcass asunder and 
dragged portions of it away. 

In Johnson's quarry the bones lay on top of each other in a heterogeneous 
mass. Perhaps he wovdd uncover the end of a limb only to find that it ran 
beneath another bone which would have to be removed before the first could be 
prepared. It was like a game of jackstraws, and only men with years of experi- 
ence and infinite patience could have done the work at all. First, the earth 
was swept away with a whisk broom; then with a camel's hair brush as the 
fossils were approached. Curv^ed awls were used to loosen the matrix. If the 
bone was soft and crumbly it was hardened with a solution of gum arabic, and 
Japanese rice-paper was stippled on every inch; thus the tiny fragments were 
held in place. After this initial treatment, the specimen was bandaged with 
strips of burlap soaked in flour paste. When the paste had dried, more of the 
earth or rock was dug away, the fossil turned over and the operations repeated 


until the entire bone had been enclosed in a hard shell. This kept it intact 
during the long journey to the laboratories of the Museum. Some specimens 
were so closely cemented together that they could not be separated in the 
field, and large blocks of bones and matrix were removed entire. 

One bed of red sandy clay near the Johnson quarry, the highest fossil- 
bearing horizon yet found in the Iren Dabasu basin, contained abundant 
fragments of smooth curved plates. We were at a loss to identify these at the 
time, but at the end of the season, when the discovery of imdoubted dinosaur 
eggs had been made at Shabarakh Usu, Granger began to suspect that these 
also represented dinosaur eggshells. Specimens were sent, together with those 
from Shabarakh Usu, to Professor Victor Van Straelen, Universite Libre, 
Brussels, Belgium, who confirmed the diagnosis, and in part remarks: "Thus 
there are striking differences between the eggs of Shabarakh Usu and those of 
Iren Dabasu. The first-named cannot be correlated with any of the actually 
known eggs, either living or fossil. The second-named have a structure similar 
to that of the supposed eggs of Hypselosaurus priscus, the dinosaurian of 
Rognac, which themselves have a structure that partakes of the characters 
of both the palsognathic and the neognathic birds. But the Iren Dabasu eggs 
differ essentially from the Rognac eggs in the shape of the aeriferous canals."^ 

The shells vary in size and probably represent different types of dino- 
saurs. The larger eggs are considerably larger and less elongate than those 
fotind at Shabarakh Usu. This is the second place in Mongolia where dinosaur 
eggs have been discovered, 


While the other palseontologists were working in the Cretaceous beds of 
Iren Dabasu, Granger and Morris made several trips to the Oligocene Houldjin 
gravels exposed at an escarpment six miles farther to the south. This is a 
rather thin capping of yellow, pebbly gravel lying on the Cretaceous, and is 
apparently only a remnant of a much more extensive formation. In 1928 
this formation was traced in the west and named the Baron Sog formation. 
It is quite certainly the residt of stream action, and the fossils are broken up 
and rounded as though they had been carried for some distance by swift- 
flowing water. 

It was at this exposure that the first fossils in Mongolia were discovered 
by the Expedition. Baluchitherium fragments were abundant, but in only a 
few cases were they worthy of being saved. Teeth and bones representing the 
following genera were also collected: Ccenopus or Prceaceratherium, Cadiirco- 

' Van Straelen, Victor. 1925. "The Microstructure of the Dinosaurian Egg-Shells from the Cretaceous 
Beds of Mongolia," Amer. Mus. Novitales, No. 173, May 27, p. 3; Preliminary Reports, Central Asiatic Expe- 
ditions, Amer. Mus. Nat. Hist., I, No. 49, pp. 1-4, 1926. 


therium, Entelodon and a large tortoise, Testiido. Dr. W. D. Matthew, who 
identified the specimens, stated that the "fauna is of OUgocene age but cannot 
be more exactly correlated until more completely known."' 


I remained at Iren Dabasu only two days, for considerable equipment 
had been left at Kalgan, which had arrived too late to go with the camels. 
Vance Johnson, Liu and I went back in two cars. On the way I had an amus- 
ing experience with brigands. My car was more than a mile in advance of 
Johnson's when we came to the spot, just below Chap Ser, where two Russian 
cars had been robbed a few weeks earlier. As I recognized the spot, I thought 
to myself, "I wonder if brigands would attempt to hold me up in this same 
place?" Hardly had the thought taken form in my mind when I saw the sun- 
flash of a gun barrel on the summit of a hill three hundred yards away. The 
head and shoulders of a single mounted horseman were just visible against 
the sky. In Mongolia and China only two kinds of natives have modem rifles 
— brigands and soldiers. As a matter of fact, they are virtually synonymous, 
for a soldier usually becomes a brigand when a favorable opportvmity offers, 
and vice versa. 

Undoubtedly, the horseman on the hilltop was a sentinel to give warning 
to others in the valley below. I had no mind to have him in such a position, 
whoever he might be, and drawing my revolver, I fired twice. The bullets 
must have come too close for comfort, although I did not attempt to hit him, 
for he instantly disappeared. A moment later, as my car topped the rim of 
the valley, I saw three mounted brigands at the bottom of the slope. It would 
have been difficult to turn the car and nin without exposing myself to close- 
range shots, and, knowing that a Mongol pony never would stand against 
the charge of a motor car, I instantly decided to attack. The cut-out was 
open and, with a smooth down-hill stretch in front of me, the car roared down 
the slope at forty miles an hour. The expected happened ! While the brigands 
were endeavoring to im-ship their rifles which were slung on their backs, their 
horses went into a series of leaps and bounds, madly bucking and rearing 
with fright, so that the men could hardly stay in their saddles. In a second the 
situation had changed! The only thing that the brigands wanted to do was 
to get away, and they fled in a panic. When last I saw them they were breaking 
all speed records on the other side of the valley. 

•Matthew, W. D., and Granger, Walter. 1923. "The Fauna of the Houldjin Gravels," Amer. Mus. 
Novilates, No. 97, p. i; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. i5, 
pp. 1-6, 1926. 



When I returned from Kalgan to Iren Dabasu I brought with me Colonel 
H. Diinlap,' Commandant of the U. S. Marine Guard Detachment, Ameri- 
can Legation, Peking, and Lieutenant-Colonel Seth Williams. These two 
officers had done much toward equipping the Expedition, and we had planned 
to have them visit us before we started west. We reached camp on May 1 1 , 
1923, after an uneventfvil trip, and for a week the officers shot sand-grouse and 
antelope and watched the excavating, at both Johnson's and Olsen's quarries, 
which was well tinder way. They left us on May 18 to return to Peking, much 
to the regret of every member of the Expedition. 


The following day. May 19, we shifted camp to Irdin Manha, twenty- 
three miles south, leaving Johnson and Kaisen to carry on their work at Iren 
Dabasu. Our tents were pitched on the edge of the bluff, near a spring which 
bubbled out of a layer of Eocene clay. To the north and west we could look 
over the rim of the basin to the sculptiired flanks of the great escarpment; 
to the south and east lay the flat reaches of the Gobi erosion plane as level as 
a gigantic polo field. 


The day after our arrival at Irdin Manha we experienced one of the worst 
windstorms that I have ever seen in Mongolia. At two o'clock in the after- 
noon a full gale was raging, and every hour the wind increased. I went out 
at four o'clock to find Granger, but I could hardly stand against the blasts 
of sand and gravel which mutilated my hands and face until they bled. The 
basin below us was "smoking" as if from a prairie fire; the yellow blanket rolled 

' In 1931, Brigadier General Dunlap was killed in France while gallantly attempting to rescue a peasant 
woman who had been buried in her cottage by a landslide. 



and swayed, now and then parting for a moment to show a bit of vegetation 
on the floor, only to have the vista closed as a fresh wave of sediment swirled 
across to the escarpment's rim. In the tents we were almost buried in sand; 
beds, clothes, tables and chairs were thickly covered with a yellow layer. 

This windstorm gave us an excellent demonstration of the methods by 
which the depression had been made. Great clouds of sand whirled and eddied 
over the edge of the escarpment; thus was the accumulated sediment carried 
out. The geologists believe it possible that a river started the excavation, 
but the enclosed basin must have been scoured out by the wind. Although 
small stream courses, dr}^ most of the year, lead from the surroimding bliiffs 
toward the salt marsh at Iren Dabasu and a certain amount of sediment must 
be transported by them into the basin, yet the depression does not fill. Only 
the wind could remove it, as we saw it doing in those gales, which continued 
for three weeks with only momentary'' lulls. It was impossible to work except 
at intervals, and then under the most trying circumstances. I drove to Iren 
Dabasu twice to see how Kaisen and Johnson were making out. They had 
stuck doggedly at it whenever there was an interval of comparative quiet, and 
had removed a great quantity of material. 


Irdin Manha, which means "The Valley of the Jewels," takes its name 
from the brilliantly polished pebbles of jasper, chalcedony, agate, quartz and 
quartzite found in the Eocene sediments. It is interesting to note that the 
pebbles were already polished when they were laid down in the deposit during 
Eocene times, but whether this was a chemical or a sand-blast polish is un- 
certain. The formation is sharply divided into two members — one of red 
clays at the base, and the other of gray sandy clays, sands and gravels above. 
It was thought wise to give the name Arshanto to the lower member, reserving 
the term Irdin Manha for the upper gray division. 


Matthew and Granger, 1926, have described two primitive little peris- 
sodactyls from the Arshanto, but, although the beds are richly fossiliferous 
in certain places, the remains are usually fragmentary. A few limb bones 
were found intact, but no associated skeletons. Teeth, pieces of jaw and 
ends of limb bones were the most usual material, although some fine skulls, 
nearly complete, were discovered. These show definitely that the formation 
is of Upper Eocene age. We did not know until later that it was near where 
the Kalgan-Urga road cuts this deposit that the Russian explorer, Obruchev,^ 

' Obruchev, V. A. 1893. "Brief Geological Sketch of the Caravan Route from Kiakhta to Kalgan." 
Bull. Imp. Riiss. Ceol. Surv., XXIX, pp. 347-390. St. Petersburg (Russian). 


,;'-'--.... '..M. 

A. RESTORATION OF Protoceratops. 

On the nesting ground of Shabarakb Usu during the Cretaceous period, by Mrs. E. R. Fulda. 

B. SKELETON OF Pfotoccratops andreiisi from shabarakh usu. 
Mounted by Peter Kaisen, American Museum of Natural History. 


A. SKULL OF Protoceratops andreu'si from shabaraku usu, 1923. 

B. SKULL OF l')0tuci:i\::ufs audrci^si pakilv lxlavatld ai shaeakaku usu, ly.2 


in 1892, collected the fragment of a "rhinoceros" jaw, which was the first 
and only fossil recorded from Mongolia, prior to the 1922-1930 work of the 
Central Asiatic Expeditions. Remains of lophiodonts were extraordinarily 
abundant, and it was possible to collect a handful of teeth in an hour, at 
almost any part of the exposure near camp. The entire absence of horses was 
a surprising feature of this and other early Tertiary faunas of Mongolia. 

The day after our arrival, Granger went to a spot west of camp to relocate 
a large jaw which he had discovered the previous year. Near the same place 
he found a beautifully preserved titanothere jaw lying across a tiny wash; 
upon starting the excavation, the corresponding jaw came into view, lying 
palate up, with a complete tooth row. The skull was almost perfect and 
there was great rejoicing in camp. A half-hour later Granger fotmd a second 
pair of jaws directly imder the first, as well as another slightly broken skull. 
That night Olsen reported that he, too, had located a fine titanothere palate. 

Our discovery of titanotheres in Mongolia was a personal triumph for 
Professor Osbom. Heretofore, titanotheres were known only from America, 
with the exception of a doubtful fragment from Austria. Just before I left 
New York, Professor Osbom said to me: "Make a careful search for titano- 
theres. I am convinced that you will find early types in Central Asia." We 
were so delighted with the discovery that I sent a cable to the Museum by a 
car which passed oiir camp en route to Kalgan. 

Titanotheres are an extinct family of peculiar, hoofed mammals, which 
flourished in North America from the Lower Eocene to the close of the Lower 
Oligocene and were represented by nearly two htmdred known species. The 
primitive Lower Eocene forms were hornless, but the climax of the group is 
represented in the Lower Oligocene by gigantic animals with a pair of trans- 
verse horns above the eyes. The Mongolian titanotheres show the most strik- 
ing resemblances to the forms of corresponding horizons in North America. 
They demonstrate clearly that there must have been a land connection between 
Asia and America, which acted as a migration route. 

During the three weeks which we spent at Irdin Manha, remains of 
twenty-six titanotheres were collected from this formation. By far the greater 
number belong to the species which Professor Osbom, 1925, named Protitano- 
therium grangeri, but other species and genera were represented, as follows: 
Manteoceras? irdinensis, Metarhinus? mongoliensis, Telmatherium berkeyi, 
Dolichorhinus olseni. 

Granger had instructed Kaisen and Johnson to finish work in their 
quarries, for they were getting only duplicates, although much bone remained 
in sight. I sent Vance Johnson, one of the motor experts, into Miao T'an, 
with two tons of fossils to be stored at the Chinese inn untU the end of the 


The afternoon of Johnson's return, "Buckshot" discovered the superb 
skull of a gigantic beast, which I believed to be a carnivore. The next day 
Granger dashed our hopes by pronouncing it to be that of a pig, Entelodon, 
which, because of its omnivorous habits, resembled a flesh-eater. However, 
Morris made a drawing of the skull in situ; this was forwarded to Doctor 
Matthew at the American Museum. When we returned from Mongolia, a 
letter was awaiting us, stating that my original supposition was correct, 
and that the specimen represented one of the primitive creodonts of the 
family Mesonychidce. 

Later it was named AndrewsarcJms mongolieiisis by Professor Osbom, 
who says that "the cranial and facial proportions of Andrewsarchiis are re- 
markably similar to those of Entelodon of the Oligocene and of Dinohyus of 
the Lower Miocene of North America, doubtless because of similar omnivorous 
feeding habits." Professor Osbom further remarks: 

"This is the largest terrestrial carnivore which has thus far been discovered 
in any part of the world. The cranium far surpasses in size that of the Alaskan 
brown bear, Ursiis gyas, the largest living carnivore, which, when full-grown, 
weighs 1,500 pounds; in length and breadth of skull, Andrewsarchus moiigoli- 
ensis is double Ursus gyas and treble the American wolf, Canis occidentalis. 
It is also treble the size of its American relative Mesotiyx obtusidens from the 
Middle Eocene of Wyoming and double that of Mesonyx {Harpagolestes) 
uintensis from the Upper Eocene of northern Utah, Uinta B." Basing his esti- 
mates upon the skeletal restorations of Mesonyx obtusidens by Scott and of 
Drontocyon vorax by Wortman, Professor Osbom says: 

"If Andrewsarchus mongoUensis was proportioned in the same manner as 
Mesonyx obtusidens, it had a length from the snout to the back of the pelvis of 
twelve feet six and a half inches, and a height from the ground to the shoulder 
or middle of the back of six feet two inches. Thus in round numbers it was three 
times the size of Drontocyon vorax or of Mesonyx obtuside7is of the Middle 
Eocene of Wyoming, Bridger formation."^ 

In addition to the specimens I have mentioned, our collections from Irdin 
Manha contained representatives of the Camivora, Rodentia, Insectivora, 
Perissodactyla, Artiodactyla and Amblypoda. The Irdin Manha formation 
was prospected for several miles both east and west, but nowhere were the fos- 
sils so abundant as in the vicinity of our camp near the Kalgan-Urga trail. In 
the western part of the basin the escarpment breaks up and the exposures are 
limited to a number of rounded hillocks. On one of these I found the upper 
premolar tooth of an animal which Professor Osbom, 1924, identified as repre- 

' Osbom, Henry Fairfield. 1925. "Andrewsarchus, Giant Mesonychid of Mongolia," Amer. Mus. 
Novitates, No. 146, pp. 1-5; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 
34. pp. 1-5. 1926 


senting the Amblypoda, an archaic order of ungulates. Except for Corypliodon 
of the Lower Eocene of France and England, no Amblypoda had hitherto been 
known in Eurasia, and their presence in Mongolia served as a new link with 
America. Diligent search revealed only this single premolar, but in September 
of 1923 Professor Osbom himself found another tooth, imder circumstances 
which will be described later. 

To the northeast, about seven miles from the road, the Gobi erosion plane 
breaks off into another broad undrained basin where there are excellent expo- 
sures, but few fossils. However, some thin beds of gray and red sandstone did 
yield a fragmentary collection of small lophiodonts, which appear to be qmte 
unlike those from the typical Irdin Manha formation. 


While working on the face of the escarpment just west of camp, the men 
were annoyed by the brown pit- viper, Agkistrodon halys intermedius, which 
was present there in numbers. When the sim began to warm the rocks near 
noon, these extremely poisonous reptiles would crawl out from their nests 
among the rocks and prevent the fossil collectors from becoming drowsy over 
their work. Each man usually accovinted for five or ten during the day. 

On the plain in front of camp, gazelles were nvimerous, and usually one or 
two herds were visible from the door of my tent. There was little difficulty in 
keeping the table well supplied with delicious meat. Several times, in the early 
morning, we saw wolves loping northward toward the edge of the basin, where 
they had dens in the face of the blviff, and they gave us some exciting races. 

The taxidermists trapped industriously dviring all of our stay at this camp, 
and collected several hundred small mammals. At Iren Dabasu they caught a 
single specimen of a new rodent, Ellobius orientalis, which Dr. G. M. Allen 
remarks "not only extends the known range of the genus to the eastern part of 
the Mongohan plateau but constitutes its easternmost record."^ The kangaroo 
rat, Allactaga mongolica, was extraordinarily abvmdant at Iren Dabasu, and in 
fact throughout the Gobi. It has a wide range and does not vary subspe- 
cifically in the desert proper. Doctor Allen said, "Among the specimens taken 
during May and early June there is a very striking preponderance of males. 
Thus, of the one htindred and seventeen skins taken in May, aU but fifteen 
were males, indicating some difference in habits in the earlier part of the sea- 
son; for in August the proportion is just reversed, with only five males to four- 
teen females."^ 

' Allen, Glover M. 1924. "Microtines Collected by the Asiatic Expeditions," Amer. Mus. Novitaies, 
No. 133, p. 13; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 28, pp. 1-13, 

' Allen, Glover M. 1925. "Jerboas from Mongolia," Amer. Mus. Novitates, No. 161, p. i; Preliminary 
Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 42, pp. 1-6, 1926. 


A beautiful three-toed jerboa, Dipus sowerbyi, was also fairly abundant, 
as were the sand-rats, Meriones auceps and Meriones unguiculatiis. Of the 
giant sand-rats, Rhombomys opimiis nigrescens, which we first discovered at 
Tsagan Nor, a single specimen was taken at Iren Dabasu, thus extending its 
range well to the east. We fovmd no representatives of the conies, Ochotona, at 
Iren Dabasu, although they occur further to the west in exactly the same type 
of country. However, the black-tailed hare, Lepus tolai iolai, was plentiful. 

The most interesting of all the birds were the sand-grouse, which were 
present at Iren Dabasu in great numbers. There were two or three small pools 
among the sand hillocks, and in the morning and afternoon thousands of sand- 
grouse came there to drink. It was a beautiful sight to see a flock sweep in, 
circle a few times, and settle almost as one bird. Their flight is exceedingly 
swift, and they furnish splendid shooting as well as food. 


Mr. Larsen had undertaken to obtain our passports for Outer Mongolia 
from the authorities in Urga, who had promised me the previous year to give 
them without trouble, but he encountered great difficulty and there were con- 
tinual delays before they were finally issued. On Monday, June ii, our pass- 
ports for Outer Mongolia were brought from Urga by a Russian car, since our 
camp was close to the main Kalgan-Urga road. We spent the day packing the 
remaining fossils and writing letters, for this would be our last contact with 
the outer world until the end of the summer. Almost a ton of specimens, 
which had accumulated since our previous consignment, was taken to Miao 
T'an and stored at the telegraph station at Iren Dabasu. We also made ready 
for an early start the next day. The caravan had already gone westward, and 
we expected to join it at Ula Usu, near the Shara Munm River, where a great 
fossil deposit had been located the year before. 


We followed a faint path westward to the small Holostai temple, and 
then turned southward on a larger trail to the great Boltai Urtu monastery at 
the jtmction of the Sair Usu road. Just before reaching this point, Morris dis- 
covered another deposit of Permian strata containing invertebrates and spent 
about two hours making a rapid investigation. On the way from Holostai to 
Boltai Urtu we passed a Mongol encampment of half a dozen yurts. The 
inhabitants were thrown into a state of terror at sight of the motor cars and 
otu^selves. After we had somewhat allayed their fears, they told us that none 
of them had ever before seen a white man or an automobile. This was most 
surprising as we were less than three htindred miles from Kalgan, although in a 
region between the main caravan trails. 


Boltai Urtu temple is beautifully situated in a deep grassy basin. The 
lamas informed us that Merin had passed that way the day before, and had 
told them that we were coming. As a result there was no panic, and we had 
hundreds of priests crowding about us to inspect the cars. From the monas- 
tery our road led through the grasslands, where thousands of female antelope, 
Procapra gutturosa, were herding just prior to giving birth to their young. 
We saw about twenty thousand does but very few bucks. At this time of the 
year the bucks lead more or less isolated lives and are seldom foimd in herds of 
more than a dozen. 

The river Shara Murun, at the time we arrived in 1922 and 1923, was only 
a thin stream of water flowing over a mud bed between low banks, and we 
crossed without much difficulty. A short distance from the rim of the western 
escarpment of the broad river valley stands the small Baron Sog monastery, 
which became a repository for specimens and gasoline of the expedition in 
1923 and 1925. The lamas are an unusually pleasant lot and were always 
most willing to assist us in any way. A fine obo stands on the very edge of 
the bluff and is visible for a long distance. During our stay we saw elaborate 
services being performed at the obo. Such celebrations at the large obos all 
over Mongolia are annual affairs. 


Our camp was close to the place where I had discovered a titanothere jaw 
the year before, and about a mile from the "Well of the Mountain Water." 
We arrived late in the afternoon, and the men could hardly control their 
impatience to inspect the new locality, which had appeared to be so rich on oiu: 
previous visit, September, 1922. The tents were pitched on a low sandy terrace 
in the basin proper, near the edge of a deeper depression where the fossils are 


The upper part of the Shara Murun formation consists chiefly of white and 
light-gray sandstones in which there are areas of gravel and beds of brownish 
and gray clay. The lower member is almost entirely sandy clay, variegated in 
red, purple, brown and green layers. Underlying the Shara Munui beds is a 
hard red clay in which a few fossils were fotmd. This appears to be a different 
formation and was named the Tuldium, from the broad lowland in which it is 
exposed. Although the Shara Murun seems to be conformable with the Irdin 
Manha farther to the east, the fossil content of the Shara Murun indicates 
that it is of somewhat later Eocene age than that of the Irdin Manha formation. 

Doctor Spock^ remarks: "During Eocene time when the Shara Murun 

' Spock, L. Erskine, III, this Series. (MS in preparation.) 


and the overlying Ulan Gochu formations were being deposited the region was 
one of level flood plains, crossed by sluggish meandering streams and dotted 
with small lakes. The alternation of red and white sediments coupled with the 
abundance of gypsum leads to the hypothesis of a climate fluctuating between 
arid and comparatively moist conditions, a state of affairs which seems to have 
continued from the Cretaceous to the present all over the Gobi Desert." 


The titanothere lower jaw, from which Granger collected the tooth row in 
great haste during one hour's halt at Ula Usu in September, 1922, was described 
by Professor Osbom, 1923, as Protitanolheriiim mongoliense. As it was the 
type of a new species, Granger wished to recover it on our return in 1923. He 
found the jaw with little difficulty and carefully pasted it with burlap. He left 
it in situ to dry, but during that time a pariah dog, which had been wandering 
about the neighborhood, was attracted by the wet flour paste and carried away 
the precious specimen. It is probably the first time that a bone millions of years 
old had furnished nourishment for a living animal! The same dog later 
damaged Albert Johnson's rhinoceros skeleton, and was making an attempt 
upon Olscn's titanothere skull when it was shot. 


The weather at Ula Usu was worse than any that we had experienced in 
Mongolia. The afternoon following our arrival, a terrific storm ushered in two 
weeks of almost continual wind. The description which I wrote m my journal 
the next day is as foUows: 

"I was making the rounds of the diggings when the strong wind which had 
been blowing all morning increased to a full gale. The basin looked like the 
crater of a volcano. Yellow clouds rolled and eddied up from the floor to sweep 
across the plain in swirling wind-devils. To the north an ominous tawny bank 
advanced upon us at race-horse speed. 

"I started into the basin to recall the men but almost instantly a thou- 
sand shrieking demons seemed to be pelting my face with sand and gravel. 
Breathing was difficult ; seeing impossible. I stumbled over the rim of the basin 
onto the plain and tried to strike diagonally toward camp. It was like pushing 
into an unresisting yellow wall which gave and closed behind me as I advanced. 
Even the ground beneath my feet was invisible. 

"In a few moments I realized that I was being carried far to the east of the 
tents. The only recovu-se was to turn into the wind tintil I found the rim of 
the basin again and crawl along it to the cut behind camp. With head com- 
pletely enveloped in my coat, I fought the salvos of sand and gravel. I do not 
know how long it was — perhaps ten minutes, perhaps half an hour — before I 


suddenly stumbled over the edge of the depression and rolled into a hollow. I 
lay there huddled against the wind trying to think. Suddenly forms took 
shape in the smother right beside me. I reached out and caught one of them by 
the leg. 

"It proved to be Tserin, one of our Mongols, with Peter Kaisen. Press- 
ing our mouths close against one another's ears we held a consultation. Tserin 
thought the tents were directly south of us; Peter and I had no idea where 
they were. I decided to trust the native's instinct, and clinging to each other 
we groped our way through the blinding murk. A few yards forward and a 
black object loomed before us. It was the cook tent, still standing, but threat- 
ening to tear in shreds at every blast. The mess tent was just beside it and we 
foimd our way inside. Lying on the floor with our faces buried in wet clothes we 
at least could breathe. 

"One by one the men blew into camp, with the exception of Walter 
Granger. It was impossible to search for him, and I was not greatly worried 
since Walter had demonstrated many times that he could take care of himself 
in any emergency, and always had a way of tiiming up, smiling. But Buck- 
shot, who worshiped Granger, was frantic with anxiety. I had to order him 
not to leave camp, else he would have dashed wildly out into the blasts of 
sand to search for his master. As Johnson remarked, 'The directions say take 
it' and we took it in whatever position we could be most comfortable. 

"The gale continued for an hour and then dropped suddenly into a flat 
calm. Not a breath of wind stirred the flag which hung limply above my tent, 
whipped almost to ribbons. The silence was vincanny after the roar and rush 
of the storm. 

"We were just crawling out of the mess tent when there was a joyous 
shout from Buckshot and we saw a brown figure coming into camp. It was 
exactly the color of the desert, but behind a broad grin was Walter Granger. 
When the storm broke he had fought his way to a partly excavated titanothere 
skioll, to mark the spot for fear it would be lost in the shifting sand. He just 
reached it, but could go no farther and huddled into the deep pit with his face 
in a coat. He had been completely buried except for his head and well-nigh 
smothered, but the walls of the pit had given him a good deal of protection. 

"We began to dig out the tents and empty the sand from our clothes and 
beds. Much of the Gobi Desert was in our belongings; it had sifted into even 
the tightest boxes. The cameras, rifles, pistols and field-glasses suffered most, 
for even their double cases coiild not keep them clean. We worked steadily for 
two hoxu's 'shoveling out.' I sent a car to the well a mile away, and everyone 
had a bath and dressed in clean clothes. Dinner was being served when one of 
the men looked out to the north. There it was again ; the same tawny cloud, 
but this time preceded by an enormous 'wind devil' which danced and swirled 


across the plain like a thing of life. It was heading directly toward us and we 
all knew what it meant if it struck our camp. I shouted for all hands to weight 
the bottoms of the tents and pound in the pegs. Explosions of wrath were 
heard from every side, because we were so clean then and knew full well how 
dirty we would be in a moment. 

"Meanwhile the sand spout continued to advance, now and then taking a 
side trip off the line to our camp, but invariably getting back again. Sud- 
denly it gave birth to a whole litter of 'wind devils' which skipped blithely away 
as though pleased to be rid of a dominating parent. 

"We awaited the inevitable, muttering curses. The attack came with a 
crash and a blast of gravel like exploding shrapnel. For five minutes it tore 
at the camp, like a wild thing trying to suck the tents and all our belongings 
into the whirling vortex above our heads. Then, repulsed at every point, it 
danced away across the plain, seeming to feel with its wraithlike yellow arms 
for easier prey. 

"Every man had a place at the poles of his own tent. Granger and I 
had held ours down together, and in the calm which followed the first attack 
we looked at each other and burst out laughing. 'Great Gods! Am I as dirty as 
you are, Roy?' he asked. I assured him that he was, only more so. When 
he looked at himself in a mirror he grunted disgustedly, 'That finishes it. The 
Mongols have the right idea, no more baths for me. What's the use? I'm 
going to bed.' The wind had begun again and developed into a full gale before 
the hour had passed." 


During the two weeks of our stay at Ula Usu there was almost continual 
wind, which made it extremely difficult to work and most imcomfortable 
merely to exist. The preceding ten rainless months had dried every scrap of 
vegetation and the Ula Usu basin showed hardly a trace of green. There was 
nothing to bind the sand and the wind whipped through our camp in unceasing 
yellow gusts. The camels arrived not long after we reached Ula Usu, and I sent 
them on to Ardyn Obo at once, as there was no grazing near camp. They were 
not in good shape, due to the lack of feed and the previous summer's hard work. 
Their hvunps were soft and flabby and many of them were decidedly weak. 


The abundance of titanothere bones at Ula Usu was amazing. There 
were many spots where hundreds of fragments lay in a white heap on the sur- 
face, the remains of skeletons which had weathered out and broken up. 
Although we discovered no complete usable skeletons, hardly a day passed that 
someone did not find a new skull or important bones. Remains of twenty- 


seven titanotheres, including two genera and three species, were collected. All 
the Mongolian titanotheres have been described by Professor Osbom, who 
remarks: "Fortunately for purposes of correlation and bearing on zoogeo- 
graphic relations in Upper Eocene and Lower Oligocene time, seven of the gen- 
era closely correspond in the two countries, namely, Mongolia and the Rocky 
Moimtain region of North America."' 

One titanothere skeleton, minus only the skull, lay in a cut bank, but the 
bones were so soft and chalk-like that it would have been exceedingly difficult 
to remove it; after being photographed and measured it was left in position. 
In addition to the titanotheres, two genera of Camivora were discovered, and 
one of Artiodactyla. The latter Doctor Matthew identified as belonging to the 
family Hypertragulidas and named ArchcBomeryx optatus; he says it "is of 
exceptional interest, as it appears to be an approximate ancestral type for the 
pecora. ... So far as the higher ruminants are concerned, it affords tangible 
and very convincing proof of the theory of an Asiatic dispersal center."^ 

In 1925, Doctor Chancy found abundant remains of fossil wood in the 
Shara Murun formation. Although some of the stems appear to have been 
large, most of them are small and twisted, resembling those of trees and bushes 
growing in arid regions with a high range of temperature. 


While the rest of us remained at Ula Usu, Morris and Young proceeded 
twenty miles to the west and camped at Jisu Honguer, where Permian strata had 
been discovered. They mapped the formation, which had been given the locality 
name, as well as recorded the rocks and collected fossils from them. This spot 
is exceedingly important, as it gave the first evidence of the great Mongolian 
geosyncline which had been so brilliantly predicted by Dr. A. W. Grabau.^ 


At the end of oxix stay at Ula Usu, we deposited more than a ton of fossils 
and a supply of gasoline at the temple Baron Sog-in-Sumu, to await otir return. 
We then moved northwestward to Ardyn Obo, over the same trail which we had 
traveled during September, 1922. Everywhere the country showed signs of the 
tinprecedented drought. There were very few Mongols, and, indeed, it was diffi- 
cult to see how any living thing coiold find sustenance on the bare brown hills. 

1 Osborn, Henry Fairfield. 1925. "Upper Eocene and Lower Oligocene Titanotheres of Mongolia," 
Amer. Mus. Novitates, No. 202, p. i; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. 
Hist., I, No. 63, pp. 1-12, 1926. 

' Matthew, W. D., and Granger, Walter. 1925. "New Mammals from the Shara Murun Eocene of 
Mongolia," Amer. Mus. Novitates, No. 196, pp. lo-ii; Preliminary Reports, Central Asiatic Expeditions, 
Amer. Mus. Nat. Hist., I, No. 59, pp. i-ii, 1926. 

2 Grabau, A. W. 1923-1924. "Stratigraphy of China," Geological Survey of China, Peking, Part I, 
p. 340, Pis. I, IV. 



Ardyn Obo proved to be an important fossil locality, as our brief survey 
of the previous year had indicated. At that time excellent skulls of the aquatic 
Cadurcotherium and fragmentary remains of new genera of Artiodactyla, 
Perissodactyla, and Camivora had been discovered. This visit yielded further 
additions to the fauna, in the shape of new genera and species in all the groups 
mentioned, as well as in the Rodentia and Amblypoda. 

Perhaps the most interesting find was the remains of three titanotheres in 
a state of evolution distinctly advanced beyond those of Irdin Manha and 
Shara Murun. One, which Professor Osbom, 1925, has named Brontops 
gobiensis, has short, broad horns and a brachy cephalic cranium. The specimen 
was discovered in the afternoon of the day before we had expected to leave 
for the Flaming Cliffs. I was anxious to get away and had announced that 
nothing but a titanothere skull would keep us longer. Therefore, when Olsen 
reported this fine specimen, just when the last packing had been done, all the 
staff were greatly amused. 


Young and I had been off on a day's prospecting trip to the southwest, 
hoping to find a trail which would take us in the direction of the Flaming 
Cliffs at Shabarakh Usu, opposite the Gurbun Saikhan range of the eastern 
Altai Mountains, without taking the long route via Sair Usu and Tuqurik 
which we had done on our return the previous autumn. A Mongol, whom we 
saw not far from Ardyn Obo, said that he knew of such a trail, and volunteered 
to guide us across country to it. We had a rough trip over hills and up long 
valleys covered with sunburned sagebrush, and eventually arrived at a spring 
called Gatun Bologai. The spot was most entrancing, a veritable oasis in pic- 
turesque surroundings. In the bottom of a bowl-shaped hollow at the base of a 
range of jagged granite mountains lay a beautiful spring, which bubbled out 
from beneath a shrinelike monument. The waters ran down the gentle slope 
in a clear stream, reflecting the emerald green of the short, thick grass which 
bordered it for many }-ards on either side. Wagtails and Mongolian skylarks 
and beautiful sheldrakes were resting on the grass, while thousands of sand- 
grouse swept in to drink. 

One can hardly realize how beautiful the oasis seemed to us after so many 
weeks without the sight of green grass! We hoped that we might bring the 
whole Expedition to see it with us, but unfortimately the guide's information 
went no further than the spring. True enough, a well-defined trail led up to 
it from the east, but there it seemed to end. We circled over the surrounding 
country in the car but could find no trace of a road leading to the west ; neither 
were any Mongols visible in the sage-covered desert. The going for the car was 


so bad that we could prospect only a small area, and reluctantly admitted that 
there seemed to be no feasible way out. Had we only known it, behind the bar- 
rier of granite peaks lay the very trail which would have taken us directly to 
our desired goal at the Flaming Cliffs! We returned upon that road in the 
autumn and were amazed to find ourselves so near the beautiful oasis. More- 
over, less than two miles to the south there was a still larger trail, a main cara- 
van route from. Kweihwating to Uliassutai! But of these things we remained 
in ignorance until the end of the summer. 


Before making this reconnaissance we had visited our caravan where it 
was encamped thirty miles south of Ardyn Obo. We interviewed several 
Mongols who had lived in the region for years, asking them about trails to the 
west. They were entirely ignorant of both these caravan trails which we foimd 
later, although they were less than sixty miles away. Later we fotind natives, 
actually living beside the trail, who had not the slightest idea where it went! 

Merin told me that it was necessary to rest and feed the camels, if they 
were not all to die within a few weeks. For many days they had been almost 
without food ; only just enough vegetation could be found on the desert to keep 
them alive. The Mongols reported the country to the west to be in even 
worse condition than that through which we had passed. It was not an 
encouraging outlook for owr caravan. 


Nevertheless, it was imperative that the camels reach the Flaming Cliffs 
with gasoline and food for us; otherwise, we would be cut off from the most 
promising fossil deposits in all Mongolia. I made up loads for sixteen camels, 
consisting chiefly of gasoline, and just enough food to carry us along on short 
rations. These were to be put on the strongest camels and pushed through at 
all costs. The other camels with their loads were to be brought on as far as 
they could go; if they died their loads were to be left. Merin felt certain that, 
under those conditions, he could at least get the sixteen loads to us at the 
Flaming Cliffs. 

We packed enough gasoline in the cars to take us to our destination, and 
food for a month. It was little enough, I felt sure, but even so the motors 
were greatly overloaded and could not carry another pound. Gasoline, food, 
and the Ardyn Obo fossils were left at the large temple near where the caravan 
was camped, to be picked up on our return. Thus we started on the long trip 
to the Flaming Cliffs via Sair Usu. 



We left Ardyn Obo on July 3, and encovintered heavy going as soon as 
we ran out of the basin onto the sandy upland. One of the cars had to be 
unloaded and its contents carried up a long, soft slope; then we used it to 
help the others through the difficult places. Several hours of hard terrain gave 
us a rest, but in the afternoon two bad river washes confronted us. The sand 
was so soft in the second that it was impossible to drive the cars on the trail, 
which ran in the river-bed for more than a mile. The rough hills, which enclosed 
the channel, at first seemed impassable, but by careful driving we took the cars 
over them, and camped tmder a large elm tree; for the first time in months 
we were lulled to sleep by the sound of wind among the leaves. 


The next morning, it took two hours of strenuous work to progress three 
miles, but that took us onto fairly good going, and we proceeded without 
incident imtil three o'clock in the afternoon, when we camped eleven miles 
from Sair Usu. It was the Fourth of July, and I did not want to let the day 
pass without some sort of celebration. Moreover, we were near the spot where 
the geologists had discovered Palaeozoic limestones; Morris wished to study 
the strata and collect its invertebrate fossils. 

Our tents were pitched on a grassy shelf which projected like a verandah 
from the base of a rugged granite ridge. From it we viewed miles of brown 
basin; just below us was a fine well, and a few himdred yards beyond it a clus- 
ter of yurts. The afternoon was devoted to "cleaning up," pitching horseshoes, 
and such other games as we could devise. A "special dinner," with a number of 
delicacies which I had provided for birthdays and celebrations, made a very 
happy ending for the evening. The next day the palaeontologists collected 
invertebrate fossils while Young and Johnson made repairs necessitated by 
an accident to one of our cars; on the 6th we pushed on again, the weather 

being clear and cool. 


Z -K 

^ =5 




This specimen was found lying directly above the first nest of dinosaur eggs discovered in 1923. Presumably it 

was an egg thief. 



We were following the Sair Usu trail, which led us over fairly good going, 
and camped at night near a temple called Menk Ta Urtu. Late in the after- 
noon we saw about two hundred Mongol recruits being drilled by Buriat 
officers, at a village of five or six yurts. As we neared them they formed in a 
double rank and stood stiffly at attention. A more ludicrous sight hardly can 
be imagined, for they had clothes of every description, and the huge boots 
with pointed upturned toes did not lend themselves to dignity when the 
wearer was on the groimd. We were told that this was a concentration camp 
for men of the district who have been drafted into the army. They are given 
preliminary instructions at this first camp, and then taken to another where 
the recruits are more advanced. Later still they go to Urga to receive training 
under Russian officers. They remain in Urga for several months, and are 
required to return every year for a stated period of training. Lamas and 
"black Mongols" alike are taken, and we heard continual complaints from the 
natives throughout the country that there were not enough men left to look 
after the herds properly. 


We picked up a Mongol guide, who offered to show us a better way to 
the Flaming Cliffs than that by which we had come the previous September, 
but very soon we found the tracks of our motor cars, made ten months before. 
However, we had the benefit of previous experience, and reached Shabarakh 
Usu on Jtxly 8 at three-fifteen in the afternoon, with little real difficulty. We 
were delighted to be there. The country through which we had traveled was 
a parched, brown desert, and even the abundant sage near the Flaming Cliffs 
was burned dry and showed hardly a trace of green. Everything was exactly 
as we had left it on our last visit. The marks of our tents and the motor car 
tracks were almost as distinct as though they had just been made. 

As soon as camp was pitched, the fossil-hunters wandered down the steep 
escarpment for a casual inspection of the new field. In less than an hour, 
Albert Johnson retiuned for his tool bag and reported the discovery of a large 
white skull. A few moments later Kaisen hurried up the slope for his collecting 
materials, and before we gathered for dinner that night every man had begun 
to excavate a dinosaur skull. Even I had had a share in the finds, for, while 
walking in the bottom of a ravine, I saw a pipe lying beside a rock. It was one 
that Granger had lost on our first visit, and strangely enough it had dropped 
within a few inches of the skull and jaws of a Protoceratops. Granger main- 
tained that he had left the pipe to mark the spot, and that I had only redis- 
covered the skull, but I insisted upon having my name painted in red ink on 
the specimen after it had been bandaged and removed. 



On July 13, George Olsen reported at tiffin that he had found some fossil 
eggs. Inasmuch as the deposit was obviously Cretaceous and too early for large 
birds, we did not take his story very seriously. We felt quite certain that his 
so-called eggs would prove to be sandstone concretions or some other geological 
phenomena. Nevertheless, we were all curious enough to go with him to 
inspect his find. We saw a small sandstone ledge, beside which were lying 
three eggs partly broken. The brown striated shell was so egglike that there 
could be no mistake. Granger finally said, "No dinosaur eggs ever have been 
found, but the reptiles probably did lay eggs. These must be dinosaur eggs. 
They can't be anj'thing else." 

The prospect was thrilling, but we would not let ourselves think of it too 
seriously, and continued to criticize the supposition from every possible stand- 
point. But finally we had to admit that "eggs are eggs," and that we could 
make them out to be nothing else. It was evident that dinosaurs did lay eggs 
and that we had discovered the first specimens known to science. 

The eggs which had broken out of the sandstone block are eight inches 
long by seven inches in circumference. They are red-brown in color and are 
rather more elongate and flattened than those of modem reptiles; they differ 
greatly in shape from the eggs of any known birds, living or fossil. The outer 
surface of the shell is striated, with broken, longitudinal rugosities, but the 
inner surface is smooth; the shell is about one millimeter thick. 


In the ledge beside which the eggs lay we could see many bits of shell 
embedded in the rock, and it was obvious that other specimens might be 
enclosed in the sandstone matrix. When Olsen brushed away the loose sedi- 
ment on top of the ledge, he exposed the fragmentary skeleton of a small 
dinosaur. It proved to represent a toothless type, and Professor Osbom subse- 
quently named it Oviraptor philoceratops. In referring to its habits he remarks: 

"The generic and specific names of this animal, Oi'i raptor, signifying the 
'egg-seizer,' philoceratops, signifying 'fondness for ceratopsian eggs,' may 
entirely mislead us as to its feeding habits and belie its character. The names 
are given because the type skull (Amer. Mus. 6517) was fovmd lying directly 
over a nest of dinosaur eggs, the one photographed being actually separated 
from the eggs by only four inches of matrix. This immediately put the animal 
lander suspicion of having been overtaken by a sandstorm in the very act of 
robbing the dinosaur egg nest."' 

' Osbom, Henry Fairfield. 1924. "Three New Theropoda, Protoceratops Zone, Central Mongolia," 
Amer. Mus. Novitates, No. 144, p. 9; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. 
Hist., I, No. 32, pp. I-I2, 1926. 


The skiill of Oviraptor is very pectiliar, being a toothless derivative of an 
originally carnivorous type, with the stout jaws much shortened in front of the 
large eyes. It was a swift-running, bipedal animal, strangely birdlike, as was 
the American dinosaur Struthiomimus. 


All the loose bits of eggshell were most carefully collected, and a large 
part of the sandstone ledge was removed en bloc and sent to the American 
Museum. Subsequently, the block was found to contain thirteen eggs in two 
layers lying with the ends pointing toward the center exactly as they had 
been left by the dinosaur when she covered them with sand for the last time, 
millions of years ago. It is probable that the nest originally contained twenty 
or more eggs, and that they were deposited in at least three layers. 

In considering the questions, why and how such delicate objects as eggs 
were so beautifully preserved, we may turn first to the geological structxire of the 
deposit. Concerning it. Professors Berkey and Morris, 1927, write as follows: 

"The rock is a red sandstone of very uniform grain and comparatively 
simple structure. The grains are exquisitely graded, so that when the stone 
is weathered, it yields a sand that will run like the sand in an hour glass. There 
is virtually no admixed clay, and separate beds of clay are few, occurring 
chiefly as channel fillings. . . . Cross-bedding, apparently of ceolian type, is 
developed on a large scale at certain horizons. We believe that the formation 
is in large part wind-blown, and that this history is the major factor in accom- 
plishing the perfect preservation of the delicate fossils which the Expedition 
recovered from it. The beds are very faintly disturbed, and show slight 
arches and saddles, but dip very gently toward the south. "^ 

Moreover, there is evidence that diuing the formation of this deposit arid 
conditions prevailed. Our palaeontologists found bits of fossilized wood in the 
beds. Doctor Chaney identified them as a desert type of tree. 

Professor Victor Van Straelen of the Universite Libre, Brussels, Belgium, 
who has studied the microstructiore of the dinosaur eggshells, remarks: 

"The eggs of Protoceratops andrewsi are of the utmost interest. From the 
rugosities of the outer surface together with the rare and extremely small 
pores, it is right to infer that the eggs had no outer cuticle. This is a character 
shown to-day by birds and turtles which lay their eggs in very dry regions. 
We may find herein a confirmation of the desert conditions prevailing in Mon- 
golia during the formation of the Djadochta beds."^ 

'Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. 157-158. 

2 Van Straelen, Victor. 1925. "The Microstructure of the Dinosaurian Egg-Shells from the Cretaceous 
Beds of Mongolia," Amer. Mus. Novitates, No. 173, p. 3; Preliminary Reports, Central Asiatic Expeditions, 
Amer. Mus. Nat. Hist., I, No. 49, pp. 1-4, 1926. 


The dry country, the loose, fine sand and high winds explain how the eggs 
were so beautifully preserved. After they had been deposited in a shallow 
depression scooped out of the sand, the dinosaur doubtless covered them with a 
thin layer of sand, and left them to be hatched by the warmth of the sun. Dur- 
ing a windstorm, such as occurs in the basin to-day, many feet of sand might 
have been deposited upon the eggs. Air and the sun's warmth were thus cut 
off, and incubation abruptly ceased. The weight of heaped-up sand even- 
tually cracked the shells, and the liquid contents of the eggs ran out. Simul- 
taneously the extremely fine sand sifted into the interiors, forming the cores of 
red sandstone which are present in all of our specimens, and prevented the eggs 
from being entirely crushed. One or two groups of small eggs which do not 
show crushing are filled with a fine-grained limy sediment, deposited from 

Since the eggs were thus gently buried in blown sand, and there is no 
evidence of water action, or of violent earth movements to disturb the deposit, 
they probably remained just as they were left, while the surrounding matrix 
became consolidated into rock. Two eggs of a group of three fovmd on the 
surface were broken, exposing the delicate skeletons of embryonic dinosaurs. 
It is probable that other eggs contain the bones of unhatched young, for we 
found several skulls of baby dinosaurs, which evidently had been out of the 
egg only a few days before death. 


In 1923, and again in 1925, we found so many eggs and remains of so many 
hundreds of dinosaurs, that it was evident that this region was a concentration 
point for the reptiles, at least during the breeding season. Geological evidence 
favors the existence of lakes to the south, where the Gurbun Saikhan now 
stands. Streams doubtless ran into them, and at least one passed through the 
"egg beds." Therefore, food and water probably were abundant. 

I believe, however, that this was not the primary reason for such a con- 
centration of dinosaurs. It appears to me that a part of the answer, at least, 
lies in the peculiar quality of the sand. For successful incubation, the covering 
sediment must be loose and porous, in order to admit air and warmth to the 
eggs. It is quite conceivable that the sand in this spot was better adapted to 
act as an incubator than that in any other part of the country. I have often 
found sand-banks which were used exclusively by turtles as a depository for 
their eggs. Other neighboring areas, where the sand was either too coarse or 
too fine, did not contain a nest. 

That this place had been a favorite breeding-grovmd for dinosaurs for a 
very long time, is shown by the fact that in 1925 eggs were found in different 
levels, all the way from the floor of the basin up to the very rim. There is a 








B. HEAD OF Oz'is UtlimOll. 


difference of two hundred feet between the highest and the lowest nests. 
Probably hiindreds of thousands of years, to say the least, had elapsed between 
the time that the eggs in the lowest and highest strata were deposited. 

A few days after the first discovery by Olsen, Granger found five eggs in a 
cluster. Albert Johnson also obtained a group of nine. Altogether, twenty- 
five eggs were collected during 1923. Some of them, as in the case of the orig- 
inal "clutch," were lying upon the surface, exposed by erosion that had worn 
away the sandstone in which they had been embedded; others were enclosed 
in the rock, with only the ends in sight. The eggs which Johnson found were 
i considerably smaller and more elongate than those of the original set, and had 
smooth outer shells. Probably they were laid by a different genus or species 
of dinosaur. 


The search for eggs was carried on coincidentally with the systematic 
prospecting of the entire basin exposures. As soon as a skull or other speci- 
men was discovered it was removed, and then the rest of the area which had 
been allotted to each man was carefully inspected. Almost every day some 
one of the paleontologists reported the discovery of new and important speci- 
mens. By far the majority belonged to the species Protoceratops andrewsi, 
which had originally been discovered by Shackelford when he found the 
deposit in 1922. The type skull is about six inches in length and belongs to an 
immature individual. It was made the type of a new family, the Protocera- 
topsidae, by Granger and Gregory, 1923, and served to show that we might 
expect to recover new forms of supreme importance from the Djadochta beds. 
About seventy skulls and twelve more or less complete skeletons represent this 
form. A series of Protoceratops skulls has been prepared, starting with an 
extremely young stage not long out of the egg and ending with a very old stage 
having a skull twenty-three inches in length. This superb series is unrivaled 
by that of any other species of dinosaur in any musetim of the world ; moreover, 
the fine sand in which the specimens were buried has preserved them so excel- 
lently that many are almost perfect. 

When the original, incomplete skull of Protoceratops was described by 
Granger and Gregory, it was supposed that the type represented a direct 
ancestor of the ceratopsians, of which the giant homed Triceratops is the most 
spectacular example. Further study of our abundant and more perfect mate- 
rial has led Doctors Gregory and Mook, 1925, to modify their views as to its 
relationship to the other ceratopsians. I cannot do better than to quote 
Doctor Gregory's own words in regard to the significance of this interesting 
and important dinosaur. He says: 

"In the second paper on Protoceratops, the present writer, in collaboration 


with Dr. Mook, showed how Protoceratops completely realizes the implications 
of Dollo's inference that even the great homed dinosaurs were secondarily- 
quadrupedal in posture, that is, that they had been derived from bipedal 
ancestors. For Protoceratops, while probably spending most of its time on all 
fours like the later ceratopsians, and while unmistakably foreshadowing the 
latter in many features, at the same time differs from them in retaining a series 
of peculiar characters in its skeleton which it has apparently inherited from 
bipedal ancestors that in many ways resembled Psittacosaurus. For example, 
its hind limbs are much longer and larger than its fore limbs, the hind foot 
being narrow and much longer than the fore foot, almost as it is in the bipedal 
PsiUacosauriis, while in the more advanced quadrupedal ceratopsians of later 
times the hind foot was much broader and little, if any, longer than the fore 

"Again, the tibia or shin bone is longer than the femur or thigh bone as in 
running, beaked dinosaurs. The pelvis is more advanced than that of Psittaco- 
saurus in the greater lengthening of the ilium, further reduction of the back- 
wardly directed pubis, initial expansion of the prepubic process and slight 
downward curvature of the elongate rod-like ischia. These and other modifi- 
cations of the pelvis indicate that Protoceratops at least spent more time 
supported on all four legs and less time running on its hind legs than did its 
less modified neighbor Psittacosaurus, while the enormous size of the head and 
jaws in proportion to the size of the backbone and thorax, as well as other 
marked specializations for the eating of herbivorous food, all give added reasons 
for the increased use of the quadrupedal posture and the gradual abandonment 
of bipedal running. Thus the Protoceratops was now getting big enough to 
stand and confront a hungry enemy, doubtless threatening him with his fierce 
beak, and no longer needed to turn and flee away on his hind legs. 

"The skull of Protoceratops shows very pronovmced modifications in the 
direction of the later ceratopsians. In the very earliest reptiles the whole 
surface of the skull behind the eyes was formed by a continuous shell of bone 
which covered over the jaw muscles. But by the time of the oldest dinosaurs 
this formerly continuous temporal region had become perforated by two 
prominent openings called the upper and lower temporal fossae. Between 
these openings were left strengthened tracts or arches, an upper middle pair, 
the parietal crest, between the upper temporal openings, a second pair, the 
postorbital arches, above and behind the eyes, and a lower pair, the jugal 
arches, just above the lower jaw. In such relatively primitive beaked dinosaurs 
as Psittacosaurus the jaw muscles were of moderate size and the temporal 
arches did not extend much behind the joint between the upper and lower 
jaws. But in Protoceratops the jaws and jaw muscles had become very large 
and in so doing had pushed their supporting arches upward, backward and out- 


ward, so that the skull is prolonged in the rear into a great spreading crest or 
frill. Formerly it was believed that this crest or frill, which is even further 
developed in the later ceratopsian dinosaurs, was evolved for the protection 
of the neck, but the construction of this region in Protoceratops plainly indicates 
that it functioned primarily as a scaffolding for greatly enlarged jaw muscles. 

"Are we to conclude then that the horned dinosatirs (ceratopsians) 
acquired their special characters in Mongolia and then spread via Manchuria 
and the Behring Straits into Alaska and thence down to the Rocky Mountain 
regions? We cannot safely affirm this on the evidence afforded by any one 
type of animals, but we can affirm that there must have been some means of 
migration either in one direction or another. Some years ago Mr. Barniam 
Brown, of the American Museum of Natural History of New York, discovered 
the fragmentary skeleton of a small homed dinosaur in the Edmonton Creta- 
ceous, Alberta, which he clearly saw was distinct from the larger homed 
dinosaur and to which he gave the name Leptoceratops. It is now evident that 
Leptoceratops is quite nearly related to Protoceratops and that it represents a 
little-changed descendant of the latter in North America at a time when the 
typical ceratopsians had reached the summit of their period of specialization. 
Some palaeontologists will undoubtedly take this fact to mean that the larger 
ceratopsians were not descended from Protoceratops but from some undis- 
covered stem form. However this may turn out, it is safe to predict that even 
if Protoceratops be not the direct ancestor of the great homed dinosavirs, it was 
at least rather closely related to that ancestor and shows us a stage in which 
the ceratopsians had just acquired secondary quadrupedal habits and an 
enlarged bony scaffolding for the jaw muscles."' 

The deep, parrot-like beak of Protoceratops may have been enclosed in a 
homy sheath. Gregory and Mook think that "the feet and tail of Protoceratops 
possibly indicate partly aquatic habits."- This may be true, but I doubt that 
the animal spent much time in the water. 


Although Protoceratops remains far outnumber all others at Shabarakh Usu, 
it was by no means the only type of dinosaur that inhabited the region during 
Cretaceous time. I have mentioned the little Oviraptor, which was foimd on 
top of the original nest of dinosaur eggs, and Professor Osbom, 1924, has 
described two other theropods from the same beds, imder the names Velo- 
ciraptor mongoliensis and Saurornithoides mongoliensis. The former, repre- 

' Gregory, William K. 1927. "Gaps in the Mongolian Life Record," The Scientific Monthly, Vol. XXIV, 
Part II, pp. 169-181. 

' Gregory, William K., and Mook, Charles C. 1925. "On Protoceratops, A Primitive Ceratopsian 
Dinosaur from the Lower Cretaceous of Mongolia," Amer. Mus. Novitates, No. 156, Feb. 11, p. 2; Preliminary 
Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., I, No. 40. 


sented by a skull and jaws, one front claw and adjoining phalanges, seems to 
have been a small, alert, swift-moving, carnivorous dinosaur. The claw of 
the first finger of its hand was very large and short, like that of a falcon. The 
latter has a long rostrum and flattened teeth. The whole aspect of the Saiiror- 
nithoides skull is so avian that we suspected that it might represent a new 
toothed bird. Professor Osbom, however, concludes that it "was a small 
cursorial theropod, more sluggish than Velociraptor, which was swift and rap- 
torial in habit, but remotely related to it."^ He thinks that it may possibly 
have been an egg-feeder, but remarks that such a conclusion must await the 
evidence afforded by the limbs. These dinosaurs are much too late in geo- 
logical time to be ancestral to birds, but they do parallel them remarkably in 
their almost winglike hands and lightly built skulls. Remains of a larger 
dinosaur were also discovered at Shabarakh Usu; as yet these have not been 


Granger discovered the skull and jaws of a small crocodile which probably 
had been an inhabitant of the stream that ran through Shabarakh Usu. Dr. C. 
C. Mook, 1924, has named it Shamostichiis djadochtcensis, remarking that it not 
only represents a new genus and species but possibly a new family as well. 


A short time after our arrival at Shabarakh Usu, McKenzie Young and I 
departed for Artsa Bogdo, sixty-five miles farther west, to hunt bighorn sheep 
and ibex. We needed specimens for both the American Museum and the Field 
Museum of Natural History. The work at Shabarakh Usu was entirely palaeon- 
tological and geological, and was so efficiently handled by Walter Granger 
that I could be of no use there. I left word that a messenger was to be sent to 
us as soon as the caravan arrived. Young and I pitched our tents near a well 
at the base of the mountain several miles west of the previous year's camp. 
It was just below the high peaks where experience had taught me that we 
would find the buck ibex which, in the summer, left the lower moimtains to the 
females and their young. 

We drove down to the spring near the trail where there was a group of 
yurts, and were enthusiastically welcomed by the Mongols, all of whom we 
had seen on our former visit. They were unreservedly glad to see us, for it 
brought much of sport and interest and money into their dull lives. Very 
quickly we concluded arrangements for the hiring of ponies and the lama 

■ Osbom, Henry FaMeld. 1924. "Three New Theropoda, Protoceratops Zone, Central Mongolia," 
Amer. Mus. NovitaUs, No. 144, p. 7; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. 
Hist., I, No. 32. 


hunter who had proved himself to be such an excellent man. Then we went 
back to camp, feeling much cheered by our friendly reception. 

The following days need not be detailed. They were a succession of 
delightful htints amid picturesque motintains. Our method was to take a led 
pony carrying food, water and blankets, and climb to the summits of the 
highest mountains. We would spend the early morning and late afternoon 
hoiars hiuiting the peaks and ridges, and either rest or shift our ground during 
the heat of the day. 

Hunting sheep and ibex at Artsa Bogdo is much less fatiguing than in 
most places, for it is possible to ride until one makes the final stalk. Having 
climbed to the summit of the mountain, we slept wherever night overtook us, 
and usually found game within a few hundred yards from where we had spent 
the night. Although we saw only a few really large heads of either sheep or ibex, 
we obtained a representative series of both species. 


There is no water high up in the mountains, so it was necessary to carry 
enough in bags to last for twenty-four hours, at least. During that time the 
ponies would not be watered at all. They were tied up at night so that they 
could not graze, as the Mongols said that it would make them thirsty to eat 
the dry grass. About noon the next day we wovdd descend to some stream or 
spring. Then the ponies would be allowed to drink their fill and graze for an 
hour or more. The Mongols have trained their ponies to go without water for 
twenty-four hours, and seldom let them drink oftener than that even when it 
is possible to do so. 

Although I had had much experience with the Mongol ponies in Peking, 
and knew their qualities, the work which they did for us on those hunts made 
me realize that the half had not been told. No true Mongol pony reaches a 
height of more than foiu-teen hands, and those of the desert, where grazing 
was poor, were seldom up to thirteen hands. Yet, they carried us up and 
down mountain slopes so steep that it was all one could do even to walk ; they 
were as sure-footed as goats, and at the end of a day of grueling climbing in 
the heat did not appear to be overtired. The Mongol pony probably represents 
a distinct type, although variations ocour in different districts, largely depend- 
ent upon feed and other conditions. It is a general opinion that he is a direct 
descendant of Prjevalski's wild horse — the differences between them being 
due to the introduction of other blood. Certain ponies show an undoubted 
Arab strain, probably being a survival of the Arab horses brought to Mongolia 
during the time of Genghis Elhan. 

These ponies are never fed grain in Mongolia. They graze the year 
round, even in winter obtaining sufficient dry vegetation to keep them alive. 


When they are brought to China they are given grain, of course, and adapt 
themselves to the change of climate, altitude, food and general living con- 
ditions in an amazingly short time. They make excellent polo and race ponies, 
and in the winter are used for cross-country hunting and steeplechasing. 
Their ability to carry a heavy man over a point-to-point course of eight or ten 
miles, with fifty or sixty jumps, compels the admiration of ever>'one who has 
seen them perform. The natives use them for racing, for every Mongol is a 
keen lover of sport. Their races are really endurance tests, for the course is 
anywhere from five to twenty miles. The ponies are ridden by boys about 
fourteen years old, who beat them from the start to the finish. 

The terrible blizzards of the Mongolian winter take a yearly toll of about 
twenty per cent of the ponies but the stronger survive and increase the endur- 
ance of the breed. The herds seek shelter in ravines and press close together 
for warmth. The hair reaches a length of five or six inches, and then the ponies 
look almost like bears. The herds are often attacked by wolves; then the 
stallions gather the mares and colts close together, and galloping aroimd them 
fight the wolves viciously with teeth and hoofs. 

In certain individuals an admixture of horse blood is present, but this can 
easily be detected when one is familiar with the breed. The head is large, the 
forehand heavy proportionately to the quarters, the chest deep, and the neck 
and legs short. Judged by horse standards the Mongolian pony is not a beau- 
tiful animal, but "handsome is as handsome does." The Mongols make no 
attempt to improve the breed of the ponies, stallions running with the herd 
and servnng the mares indiscriminately. 


While Young and I were at Artsa Bogdo, our minds were not entirely at 
rest, for the expected messenger announcing the arrival of the caravan did not 
come. We became so worried at last that I decided to leave the cook and 
camp-gear at Artsa Bogdo, put all the gasoline into one car, and drive back to 
Shabarakh Usu. 

We reached there without incident on July 30, but were disappointed to 
learn that there was no word of the caravan. Matters were becoming serious, 
because there was only a very little gasoline left, and food and other supplies 
were nmning short. I immediately rationed our remaining food, and reserved 
all the gasoline for emergency use in a single car. Since the flour was almost 
gone, by unanimous vote it was kept only for pasting fossils. We were reduced 
to tea and meat. Fortxonately there were many antelopes, and we purchased 
several sheep from the Mongols. We got along well enough, oiu- only real 
hardship being the lack of sugar. 

I also sent out our Mongols to the north and south on horseback to see if 


they could get any word of our camels from the resident natives. Thej^ all 
returned without information. At last I selected two of oiir most trustworthy 
Mongols, Tserin and Bato, and instructed them to go back eastward on two 
different routes continuing until they foimd the caravan or arrived at the place 
from which it had started. The trails eventually converged, and Bato returned 
while Tserin continued eastward. He rode ponies tintil he reached a point 
where the feed was too scanty for horses, and then hired a camel. After 
riding six or seven days across an arid waste without seeing a human being, he 
encoiontered two lamas who attacked and robbed him. He was beaten insensi- 
ble and was very nearly killed. Fortunately, the attack occurred only a short 
distance from a temple; he made his way there and remained for some time 
tmtil he was able to start back to us. A month later he reached oirr camp in 
very bad condition, having ridden or walked more than three hundred miles. 
Almost anyone but a Mongol would have died. The poor fellow was heart- 
broken because, since he was so ill that he could hardly ride, he had returned 
without fvdfilling his mission. 

One day a wizened old lama priest rode into camp. Our Mongols greeted 
him with the greatest reverence and told us that he was a famous astrologer, 
who had heard of our predicament and had come more than thirty miles to 
help us. The Mongols said that he would be able to tell us exactly where the 
caravan was. The old fellow made elaborate preparations and, after a long 
incantation, announced that the caravan was many days' travel away from us, 
but that we would hear definite news of it in three days. He said that our 
camels were dying and that Merin was having a very difficult time. Our Mon- 
gols believed him implicitly. As a matter of fact, we did hear news of Merin 
in four days, because one of my men discovered him sixty-five miles to the west 
of us, at Artsa Bogdo, which was the destination that I had given him. He 
had found it impossible to cross the sim-parched desert and had circled far 
to the north, where there was better feed, leaving his camels at wells along the 
trail, or wherever they died or became so weak that they could no longer 
travel. Out of the seventy-five camels, sixteen came through, carrying food 
and gasoline and, above all, sugar! Eventually twenty-three more reached 
Artsa Bogdo. They had been left at a well in charge of two Mongols, and had 
been able to find siifficient food to give them strength to go on slowly. To 
celebrate the arrival of the caravan we had a big dinner with camel sage for 
table decorations, August 9. 


During the time that Young and I were at Artsa Bogdo, Morris had 
roamed the country on camel-back, studying geological problems. During his 
wanderings he discovered a new and exceedingly important formation in the 


extreme eastern end of the Shabarakh Usu basin. This was an exposure of 
Paleoccne age, the very earliest period of the Age of Mammals; it lies uncon- 
formably upon the Cretaceous deposits of the Djadochta formation which 
contain the dinosaur eggs and Protoceratops. The Djadochta evidently was 
subjected to a long period of erosion before these new Tertiary sediments were 
deposited upon it. About two hundred feet of gravel, and red and brown 
sandy clays, compose the formation, which was named Gashato. 

Morris, Olsen and Buckshot spent some time searching for fossils in the 
exposure, but found bones in only two or three small pockets. Nevertheless, 
these few remains represent seven mammalian orders. In many respects it is 
a most puzzling fauna. In addition to two genera of the primitive multituber- 
culates, there are parts of small upper and lower jaws about an inch long, bear- 
ing both molar and premolar teeth. These specimens Dr. W. D. Matthew and 
Walter Granger have named Palccostylops ituriis. The genera have very definite 
affinities to the strange little hoofed mammals of Patagonia, called notoungu- 
lates. A tiny lower jaw from the Lower Eocene of Wyoming was the only pre- 
viously known fossil of this type, except the South American specimens. Thus, 
Doctor Matthew remarked: "The minute and primitive notoungulate PalcEos- 
tylops confirms the view that the South American Tertiary hoofed mammals 
were originally derived from the north, although imdergoing a great secondary 
evolution in the Neotropical region."* 

On the other hand, the fauna as a whole is somewhat surprising, as it does 
not represent the ancestral relations to the Eocene fauna of Europe and 
America that had been anticipated. 

In 1925, a further collection was made from this formation, which included 
representatives of one new family, five new genera and six new species. Con- 
cerning it Doctor Matthew and his colleagues write: "Most of the mammals 
of the Gashato appear to represent phyla previously unknown. From the 
standpoint of European and American early Tertiary mammals they are 
aberrant, and, despite their early age and primitive stamp, they are for the 
most part too speciaUzed and peculiar to cast much light either on phylogeny 
or on correlation."^ 


Almost immediately upon arrival of the caravan, Olsen and Buckshot 
began to pack the great pile of fossils that had accumulated in the tents. The 

' Matthew, W. D., and Granger, Walter. 1925. "Fauna and Correlation of the Gashato Formation of 
Mongolia," Amer. Mus. NovitaUs, No. 189, p. 2; Preliminary Reports, Central Asiatic Expeditions, Amer. 
Mus. Nat. Hist., I, No. 56, pp. 1-12, 1926. 

2 Matthew, W. D., Granger, Walter, and Simpson, George Gaylord. 1929. "Additions to the Fauna 
of the Gashato Formation of Mongolia," Amer. Mus. NovitaUs, No. 376, p. 2; Preliminary Reports, Central 
Asiatic Expeditions, Amer. Mus. Nat. Hist., II, No. 93, pp. 1-12, 1930. 


proper care of delicate specimens for their long journey across the desert was 
one of the greatest problems of the Expedition, for there is no wood of any kind 
in the Gobi and no other packing material than stiff grass. The food and 
gasoline cases provided boxes. Whenever the cars met the caravan we took 
food and gasoline from the wooden boxes and substituted fossils and other col- 
lections. The packing material was obtained from the animals themselves. 
The Mongolian camels grow very long wool to protect their bodies during 
the bitter cold of the winter, and, as the weather becomes warmer, this coat 
falls away in strips and patches. Whenever we wanted to pack a box, we sim- 
ply pulled the necessary quantity of wool off our camels. No finer packing 
material could be devised, and a new crop continually appeared as the weather 
grew warmer and the camels shed more readily. But a certain amoimt of 
care had to be exercised in plucking the beasts; for a camel, in spite of his size, 
is a very delicate animal. If we removed his underclothes too suddenly he 
would very likely catch cold and whimper in the most disconsolate way, while 
great tears ran out of his eyes. 


Just before we left camp at the red beds. Granger, Morris and I drove to 
the Gurbun Saikhan, "The Three Good Ones," an isolated range of the 
eastern Altai Mountains. It was August 10, and a day that I shall always 
remember because of the strange haze that hung over the desert. The year 
before, Berkey and Morris had explored the western end of the Gurbun Saikhan 
on camels, but had not gone to the north and east. 

Well over toward the moiuitains we had an amazing spectacle of wild 
life — the largest herd of antelopes I had ever seen. The entire horizon appeared 
to be a moving line of yellow bodies and curving necks. As we ran toward them 
in the car, the great herd divided into groups of bucks, does and yoimg. Thou- 
sands passed in front of us, sometimes stopping to gaze curiously at the car, or 
running just fast enough to keep at what they thought was a safe distance. 

Nowhere else, except in Africa, wotild it be possible to see such a herd of 
wild animals. We estimated that at least six thousand were immediately in 
front of us, but there may have been twice that nimiber, for the yellow groups 
stretched far beyond oior sight. They were feeding upon rich grass along the 
lower slopes of the Gurbtin Saikhan, where the mountains insured a greater 
rainfall. They belonged to the short-tailed species, Procapra guUurosa, which 
lives only on the grasslands. 


We were ready to leave the Flaming Cliffs on August 12. Even though 
we had been there for five weeks, specimens were still being discovered and 


each one seemed finer than the last. Kaisen found a beautiful skeleton of 
Protoceratops, nearly complete, just before we left. It was lying on its belly, 
head out, with all four legs drawn up as if ready for a spring. Apparently the 
animal had not been moved since it dropped there in death millions of years 
ago. It was too fine a thing to leave, even though I was anxious to get away, 
and I told Kaisen we would wait while he took it out. But three others, which 
Olsen and Buckshot discovered, were left untouched. We had to stop some- 
where, for apparently there was an inexhaustible supply of specimens in the 
wonderfiol basin. From that one locality our collection niunbered sixty cases 
of fossils, weighing five tons. It included seventy skulls, fourteen skeletons 
and twenty-five of the first dinosaur eggs ever seen by human eyes. As Granger 
and I in departing gazed upon the glorious spires and battlements of the Flam- 
ing Cliffs, we felt that the desert had paid its debt. 



On August 12, the Expedition moved westward from Shabarakh Usu 
bound for the Oshih basin, the fossil-bearing locality opposite Artsa Bogdo, 
which Granger had discovered the previous August. Liu, one of the Chinese 
assistants, was left with a great pile of specimens to be brought up the next 
day. We looked forward to another harvest at Oshih, for during the few days 
Granger had spent there he had removed a beautiful and complete skeleton 
of a primitive dinosaur, Psittacosaurus mongoliensis, and had done but little 

From the trail opposite Artsa Bogdo, we made our way across country 
directly north to a narrow gateway through the rocks, and came down into a 
wide, almost flat lowland. A great red mesa, nmning east and west and 
capped with black lava, occupied the center of the basin. It was a most spec- 
tacular formation, and illustrated beautifully how the thick covering of lava 
had protected the mesa from the erosion which had worn down the svirround- 
ing sediments. 


Across the western end of the mesa is a low wall of lava blocks, which 
extends for a mile or more weU out into the lowland on either side. That the 
wall is very ancient was evident, but what its purpose could have been we were 
at a loss to know. Unless it had been extended right across the basin, thus 
blocking off the mesa and the entire western end, it coiald not have served as a 
means of defense. The Mongols could give us no information, merely saying 
that it had been "made by a people who lived a long time ago before the Mon- 
gols came." 


A dry stream-bed ran along the base of the mesa and made an excellent 
road for our cars. For five or six miles we went eastward, between the mesa 


and the rocky hills to the south. Suddenly, without warning, we came to the 
brink of a vast complex of ravines, gullies and canyons, where the floor of the 
basin has been broken up by erosion into one of the most amazing and spectac- 
ular amphitheaters that I have ever seen. We pitched the tents on the very 
edge of the fantastic chasm. The wild chaos of scvilptured walls and winding, 
jagged ravines suffused an atmosphere of unreality about our camp ; it seemed 
that we were living in the world of yesterday, and that at any moment huge 
dinosaurs might wander to the doorways of our tents from out of the vast red 
canyons. Concerning the formation of this interesting region, the geologists, 
Berkey and Morris, 1927, have said: 

"The depression known as Oshih is not a structural basin, but is essentially 
a desert hollow about eight miles wide from north to south, and at least fifteen 
miles long from east to west. The erosion features, therefore, are particularly 
impressive. The high colors of the strata and the badland dissection stand 
in striking contrast to the simple features of the surrounding rolling plains. 
A red mesa about three miles long, capped with black lava, stands in the 
middle of the hollow. It is a striking landmark. At the eastern end of the mesa 
one enters a great amphitheater-like depression, extending about three miles 
to the top of the hill Oshih Nuru, which forms the eastern end of the Oshih 
hollow. The sides of the amphitheater are fretted into a splendid badland 
area. From the north wall of the hollow, a great bluff about ten miles long, 
called Urulji Nuru, one can look northward over a perfectly simple, almost 
level topography representing the Gobi erosion plane; but in the depression 
everything is minutely dissected down almost to the very bottom. The 
myriad gullies, joining into fair-sized canyons which open into smooth floored 
valleys, indicate that the dominant erosive agent in the hollow is running 


The lowest member of the strata in the Oshih basin is a red sandstone, 
which contains a great number of spherical concretions. They vary in diame- 
ter from an inch or two to several feet, and cover the bottom of the basin as 
though it had been bombarded with stone cannon-balls. It was in this red 
layer that the Psittacosaurus skeleton was found. 


Although Granger's first work in the Oshih basin had given promise of a 
rich deposit, it soon became evident that owe expectations were not to be ful- 
filled. By good fortune, he had found the only well-preserved specimen which 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, pp. 268-269. 





September i6, 1923. Irdin Manila Horizon. Pit which yielded the first perfect sk.ill. in the month of 

May, 1923. 



was exposed. Diligent search by all the members of the Expedition failed to 
produce anything like the adult Psittacosaiirus skeleton and eventually we had 
to mark the Oshih basin as the single locality of the 1922 discoveries which 
had disappointed us. 

A good deal of fragmentary material was collected. From this, two good 
skulls of baby Psittacosaiirus were developed in the American Museum labora- 
tory, as well as a fine adult skull and jaws in a large nodule. 

Fragmentary specimens of other dinosaurs were also discovered. Two 
teeth, badly weathered limb bones, dorsal vertebrse, ribs and chevrons of a 
giant sauropod, which Professor Osbom, 1924, has named Asiatosaunis mon- 
goliensis, gave the first indication of these huge dinosaurs in central Asia. He 
believes that it is related to Camarasaurus, but the material is too incomplete 
to make any precise comparison. The presence of a carnivorous dinosaur was 
also denoted by two teeth; this species has been designated as Prodeinodon 
niongoliensis . The skeleton of a small dinosaur, in a block of hematite, was 
discovered by Granger, but the matrix was so hard that the strongest tools 
were blunted. Since it wovdd have been almost impossible to prepare it in 
any case, it was regretfvdly left in situ. 


Although we were disappointed in the palaeontological results at the Oshih 
basin, nevertheless our stay of nine days was interesting. Twice we saw big- 
horn sheep on the hills, and one day, while sitting in camp, I saw a splendid 
ram standing on a pinnacle of rock projecting from the mesa; every night we 
heard the mournful howl of wolves and the sharp bark of foxes, deep down 
among the tortuous ravines. 

Young, Vance Johnson and I ran back to the trail one day, and westward 
along it, looking for wild ass. We discovered several herds, and Young and 
Johnson each shot one specimen for our collection. As we were returning, about 
forty wild asses appeared from a valley between two small mountains. In 
close formation they swimg past the car like a troop of cavalry, keeping per- 
fect alignment imtil they disappeared over a low rise in the ground. 


On August 21, we left the Oshih basin for a few days' shooting at Artsa 
Bogdo before turning homeward. A new cap of snow covered the summit of 
Baga Bogdo, and the air was already tinged with autumnal sharpness. Half- 
way to the trail we stopped to gather geodes, quantities of which were scat- 
tered among the lava blocks. 

I had promised the men a real vacation when we reached Artsa Bogdo. 
We pitched camp high above the plain, at the very foot of the mountains, near 


a well of excellent water. Ponies and hunters were hired from our friends in 
the Mongol village, and every man except Vance Johnson killed a sheep or an 
ibex during the allotted three days. On August 24, we broke camp hurriedly 
to avoid a torrential rainstorm which was gathering over the mountain sum- 
mit. Rain caught us before we had driven the ten miles to the trail, where the 
caravan was camped. We got only the outer edge of the storm, but the roar of 
water on the slopes of Artsa Bogdo could be heard throughout the night. 


All our fossils and every item of unnecessary equipment were given to the 
caravan. Nevertheless, the cars were greatly overloaded with gasoline and 
food when we started eastward on August 25, at eight-thirty in the morning. 
An accident to one of the cars forced an early camp, near some pools of dirty 
rainwater in a dry river-bed. We named the spot "Clutch Collar Camp," 
because a similar breakage had occurred on the same spot the previous year. 
Next day one of the pinion gears broke on Number Five truck, when we were 
less than five miles beyond our night's stop. The day's run netted only sixty- 
five miles. 

Our destination was Toylee-in-Sumu, where we had left the fossils obtained 
at Ardyn Obo as well as a supply of gasoline. How to get there was somewhat 
of a problem. Our hope was to discover a trail that led northward, but the 
few Mongols whom we saw maintained that none existed. On August 27 we 
made an attempt to run northward across country, but sand and extremely 
bad going drove us back to the trail. That night we camped in a wide, drj' 
wash, which led into a narrow canyon through a low rocky ridge of granite. 
Half a dozen elm trees in the dry bed gave us a delightful surprise, and wood 
for camp-fires. 

The next morning, while sitting at breakfast, I looked out of the tent door 
to see a herd of bighorn sheep feeding quietly at the upper end of the canyon. 
Vance Johnson immediately went out with his rifle and killed a young male. 
He was the only one of the party that had not shot a sheep or ibex at Artsa 
Bogdo. I was surprised to find sheep at such a low elevation, but we discovered 
subsequently that they inhabit isolated ridges far to the eastward. They make 
long journeys over the flat desert to reach high ground. 

After driving forty-two miles eastward on a rough trail, we ttimed north 
across country to three Mongol yurts which were visible in the distance. The 
natives informed us that "not far away" was a good trail which led to Gatun 
Bologai, the beautiful spring and oasis which Young and I had discovered on 
the outward journey. If we could reach that place, we should know exactly 
where we were. A native from the yurts volimteered to guide us to the trail. 
His "not far" proved to be fifty-four miles, across some of the worst going for 


heavily loaded cars that it had ever been my misfortune to encounter. Worn 
in body and spirit, we reached a broad trail eventually. Continued heavy 
sand forced us to camp at dark, just behind the granite ridge which backs the 
oasis on the west. 

We reached Gatun Bologai next morning. Much to our disappointment, 
half a dozen caravans were encamped in the immediate vicinity, and the 
beautiftd green lawn was a mass of sticky mud. Young and I had to endure 
the jokes of the rest of the party about the oasis of which we had drawn such 
alluring pictures. But the rest of the country was amazingly changed for the 
better, due to several rains which had tinged with green the parched desert 
vegetation. At Gatun Bologai we were in known country, and arrived at 
Toylee-in-Sumu by one-thirty of the same day. After taking aboard the 
gasoline, food and fossils which had been stored in the temple, we continued on 
to Ula Usu. 


According to plans made before we left New York in the spring. Professor 
and Mrs. Osbom were to arrive in Peking the first week in September. I 
decided to go in at once, with Young, to bring Professor Osbom up to the fossil 
fields at Irdin Manha. Vance Johnson and a Chinese chauffeur accompanied 
us with two trucks, to take the fossils stored at Baron Sog-in-Sumu to Erhlien. 
Granger was to remain at Ula Usu as long as he saw fit, but to move camp to 
Irdin Manha before September 15, when I expected to return with Professor 

We reached Erhlien (Iren Dabasu) on September i, without incident 
except a broken axle, and Young and I started the same day for Kalgan. John- 
son and the Chinese returned to Ula Usu with their two cars, after storing the 
fossils in the Erhlien telegraph office. Later the fossils were brought by car to 
Kalgan. Badly worn tires gave us considerable difficulty on the way, but we 
arrived at Kalgan on September 2, and I took the train the next day for 


I never shall forget my shock, on being met at the station by Mrs. 
Andrews, Colonel Dunlap and Colonel Williams, to hear the news of the Jap- 
anese earthquake. Professor and Mrs. Osbom were aboard the S.S. President 
Madison, which was due to leave Yokohama on the day that the earthquake 
occurred. I had suggested that they take the train to Kobe instead of going 
by ship. If they had done so, it was quite possible that they had met death 
in the disaster. For three days, without success, we made every effort to 
learn what had become of the President Madison and other ships which were 


known to have been in the harbor. The Japanese Government, fearful of 
attacks by other nations if the extent of the catastrophe became known, had 
instructed their war vessels to operate their wireless sets in such a way as to 
prevent radio communication. Such a mistaken conception of the sporting 
and humanitarian instincts of western nations caused much unnecessary delay 
in giving assistance to the sufferers, and untold worry to those who had friends 
or relations in Japan. Fortunately, the President Madison had left Yoko- 
hama and was sailing into the bay at Kobe when the earthquake occurred. 
Professor and Mrs. Osbom knew nothing about it imtil they were greeted in 
Shanghai by newspaper correspondents, and telegrams from the Secretary of 
State and myself. 


On the ninth of September, Mrs. Andrews and I met Professor and Mrs. 
Osbom in Peking. I was so enthusiastic about the discovery of the dinosaur 
eggs and our other great finds, that, in spite of my resolutions, I could not even 
wait until we had reached home, but told the story to Mrs. Osbom before we 
had left the station platform. The Professor was so anxious to join the Expedi- 
tion in Mongolia that we left for Kalgan two days later. The trip up the Pass 
was all that we had wished, and the night at Miao T'an gave Professor Osbom 
his introduction to a Chinese inn. We could hardly have had more beautiful 
weather, and the Professor's enjoyment, in being at last in the land of his 
prophetic vision, was immensely stimulating. 

We reached Irdin Manha at four o'clock in the afternoon. Far in the dis- 
tance, the blue tents showed in a beautiful mirage which waved and danced 
about in the sky, and settled to earth only as we drove into camp. After tea we 
walked to the fossil bec^s, where a titanothere jaw had been partly exposed for 
Professor Osbom's inspection, and then we visited the pit from which our first 
complete skull and jaws of a titanothere had been taken that spring. (PI. 
LVII, fig. A.) It was within a few himdred yards of this same spot that Granger 
had discovered the first Asiatic titanothercs in 1922. We spent the next day 
at the fossil deposits of Iren Dabasu, which have gone into history as the first 
identified Cretaceous strata and dinosaur remains on the central Asian plateau. 


Professor Osbom was exceedingly interested in a specimen which I had 
found at Irdin Manha. It was a single premolar tooth, representing an archaic 
group of mammals known as the Amblypoda. None of these great ungulates 
had hitherto been known in Eurasia, excepting Coryphodon of the Lower 
Eocene of France and England. This single upper premolar tooth was the only 
specimen of the group we had discovered in two years' search. Professor 


Osbom considered it so important that he asked to be taken to the bench, 
about two miles from camp, and to have me photographed on the spot where I 
had picked up the tooth. 

Later we drove ten miles down the valley and stopped for tiffin. As we 
were returning, Professor Osbom pointed to a low, sandy exposure a half mile 
away, and said: 

"Have you prospected that knoll?" 

"No," I said, "it is the only one in the basin that we have not examined. 
It seemed too small to bother about." 

"I don't know why," said the Professor, "but I would like to have a look 
at it. Do you mind running over?" 

When we stopped at the base of the hillock, I did not leave the car, but 
Professor Osbom and Granger walked out to examine the exposiu-e. As he 
left, the Professor turned to me and said with a smile: 

"I am going to find another coryphodont tooth." 

Two minutes later he waved his arms and shouted, "I have it — another 

I could hardly believe my eyes and ears. Jumping out of the car, I ran 
to the spot. The tooth that I had found was the third or fourth upper premolar 
of the right side. The one Professor Osbom had discovered was the third or 
fourth upper premolar of the left side, and of almost exactly the sam.e size. 
Naturally they could not have been from the same specimen, as the two had 
been fotmd eight miles apart. These are the facts. The explanation of this 
remarkable telepathic coincidence is left to the psychologist. 


One afternoon at Irdin Manha we had an amazing experience with a wolf. 
Professor Osbom wished greatly to see for himself how fast an antelope cotild 
run. From camp, half a dozen gazelles, GazeUa subgutturosa hiUieriana, were 
visible on the smooth erosion plane. In the toiuing car, we raced them for a 
few miles and finally singled out a fine buck. He loped along at forty miles an 
hour, easily keeping in front of us, and would go no faster. I told the Professor 
that the only way to make him speed up was to drop a few bullets close beside 
him. As I stopped the car to shoot, a wolf suddenly dashed from behind the 
rim of a ravine and tried to catch the gazelle. Evidently it had been lying in 
wait, watching the buck, and when it saw that the gazelle would pass no closer, 
had made his dash. The wolf had almost reached the gazelle before the latter 
saw the danger; with a few boimds the gazelle leaped to a speed of sixty miles 
an hour, hopelessly outdistancing the pursuer. It was a remarkable demon- 
stration of the method of attack by wolves upon antelope, and particularly 
surprising because of the conditions under which it was made. In spite of the 


roaring motor car and half a dozen shots which had been fired, the wolf 
still made the attack. One would have supposed that the animal would have 
been so frightened that it would have slunk away, or remained concealed in 
the ravine, no matter how hungry it was. 


The last night before the Expedition returned to Kalgan we camped in a 
beautiful amphitheater among the grass-covered hills. I did not know tmtil 
afterward that the Professor was timing the operation of making camp. It 
was exactly fovirteen minutes from the time that the last car stopped in its 
place in the line until all the tents were up and the cook-fires burning. Every 
man knew his work so thoroughly that not a moment was wasted. 


Professor Osbom and I sat for a long time in front of his tent, discussing 
the future of the Mongolian explorations. We had opened a new country 
which had given vmdreamed-of revelations in the prehistory of the earth. It 
was obvious that the work could not be concluded satisfactorily in the five years 
originally allotted for the Expeditions. Eight years we estimated to be the 
minimum, and our decision was made before the Mongolian twilight had faded 
into darkness. 

I felt strongly that the entire scientific staff ought to return to America to 
make a preliminary study of the collections and to obtain a new perspective on 
the work. Future explorations could be made much more intelligently if we 
knew what new facts our collections and data were bringing to light. More- 
over, all of us had been so close to the work for three years that we felt the 
necessity of viewing it from a distance and through eyes other than our own. 
Of coiu"se, the increased time needed for the Expeditions meant almost double 
the original financial support. To obtain another quarter of a million dollars 
would be my chief business during the coming year. 


I immediately decided to sell all our motors in Kalgan. Vance Johnson 
was left to supervise this business, and in three days every car had been dis- 
posed of at amazingly good prices. Fur dealers and wool buyers had been 
watching our explorations with the keenest interest. Many had come to us at 
the end of the first year to inquire about routes for reaching the interior of 
Mongolia. We had told them how they could go and where to send gasoline 
for their prospective trips. Several cars had followed our advice with success. 
It was an evidence of how rapidly commerce follows on the heels of exploration. 
The dealers wanted our old cars, rather than new ones of the same make. 


They had made the journey twice, and the Chinese felt that others might not 
be as good. As a result, we actually sold our cars for more than I had paid 
for them in America. 


After returning to Peking, Professor Henry Fairfield Osbom had several 
conferences with the Premier, Minister of Foreign Affairs and other prominent 
Chinese who were interested in the plan to form a Chinese National Museum 
of Natural History. I had already done considerable work on the project, 
and a committee had been formed comprising most of the Cabinet officials. By 
Presidential mandate, a section of the superb buildings in the Forbidden City 
had been designated as a home for the Museum. The Central Asiatic Expedi- 
tions and the American Museum of Natural History offered to give duplicate 
specimens of all Mongolian collections. Also, a complete set of the American 
Museum publications was promised by Professor Osbom. 

The project was making rapid progress toward a practical beginning, 
when one of the usual Chinese wars sent the Cabinet fleeing for safety to the 
Foreign Concessions at Tientsin, and nullified all our endeavors. Neverthe- 
less, a large collection of Mongolian duplicates was deposited in the Museum of 
the Geological Survey, Peking, together with a complete set of American 
Museum publications. 


Merin had made a record return journey from Artsa Bogdo, bringing 
back all our collections and equipment in safety. We had the satisfaction of 
seeing them in Peking before I left for New York in mid-October with Professor 
and Mrs. Osbom. Mr. Granger remained to pack and ship the collections and 
to bring with him two of our Chinese assistants, Liu Hsi-ku and Kan Chuen- 
pao, "Buckshot." These men had done such excellent work in collecting fossils 
that we felt they were worthy of a year's education in the laboratories of the 
American Museum. I am glad to say that they fully justified our confidence. 
The remainder of the scientific staff had sailed for America before I left. The 
camels spent the winter and the following summer grazing on the plateau 
near our Mongol village, a hundred and twenty miles north of Kalgan. 



I DID not realize what enormous popular interest had been aroused by 
ovir discovery of dinosaur eggs, until we arrived in Seattle. Newspaper repre- 
sentatives boarded the ship at Victoria, each one authorized to make a cash 
oflfer for the exclusive use of the egg photographs. Some of these reached 
thousands of dollars. I refused them all, however, as I could make no pub- 
licity arrangements until the Editor of Asia Magazine had made his selection 
of photographs with which to illustrate the popular articles I was under contract 
to write for them. Eventually the photographs were given to the press without 

The world-wide interest in the discovery of the dinosaur eggs assisted 
enormously in the financial campaign which I had to face during the winter 
of 1923-1924. Purely for the sake of publicity, we decided to sell a single egg 
to the highest bidder, the proceeds to go to the Expedition. This directed 
attention to the fact that we were urgently in need of funds with which to 
continue our work. Oflfers for the egg were received from England, France, 
Australia, New Zealand and many parts of the United States. It was finally 
purchased for five thousand dollars by the late Colonel Austin Colgate, and 
presented to Colgate University. 

Although the sale of this egg aided enormously in stimulating public inter- 
est in the Expedition, nevertheless, I have wished many times since that the 
sale never had been made. It gave the impression to many Chinese, and to the 
Russian and Mongol authorities in Urga, that the fossil collections which we 
had gathered were of enormous commercial value. They naturally believed 
that all the eggs obtained were worth five thousand dollars. They never could 
be made to understand that that was a purely fictitious price, based upon care- 
fully prepared publicity ; that actually the eggs had no commercial value. As a 
matter of fact, we did try to sell another later and found no market at any- 
thing like five thousand dollars. Beautifully executed plaster casts of the eggs 
had been made and distributed gratis to the principal world museums, and their 

directors considered them quite good enough for exhibition purposes. 



A winter of strenuous endeavor on my part resulted in the accumulation 
of two hundred and eighty-four thousand dollars for the Expedition. It came 
from almost every state in the Union, and conclusively demonstrated the wide 
public interest in our Mongolian work. 


The scientific staff made excellent progress in studying the material gath- 
ered during 1922 and 1923. Professors Berkey and Morris were busily engaged 
upon Voltune II, the "Geology of Mongolia," which was to be the first volume 
of the final publications of the Expedition. Mr. J. T. Nichols began the study 
of the superb fish collections which had been sent back by Mr. Pope. The 
reptiles collected by Mr. Pope had been accepted for determination by Mr. 
Karl Schmidt, of the Field Museum of Natural History, Chicago, by arrange- 
ment with President Stanley Field. The Field Museum had become a con- 
tributor to the Expedition and was to receive a duplicate set of all our collec- 
tions, so far as possible. 

Dr. Glover M. AUen, of the Museum of Comparative Zoology, Harvard 
University, agreed to identify and describe the ten thousand or more mam- 
mals which had been collected by the Asiatic Expeditions. In palceontology. 
Professor Osbom had taken up the study of certain groups; Doctors Matthew, 
Gregory and Granger were busy on others. New species, and such discoveries 
as it was deemed wise to put on record as soon as possible, were published in the 
American Museum Novitates or Bulletin. A steady stream of papers was thus 
issuing from the Expedition. 


I made an arrangement with the Dodge Brothers Corporation for a fleet 
of new cars, especially designed for our purposes in the desert. The changes 
which we desired consisted of heavier springs, increased radiation, larger gas 
tanks and open "express" bodies. The cars which they produced were so satis- 
factory that we could suggest no improvements, even at the end of the 


When considering the 1925 Expedition we were unanimous in agreeing 
that certain additions should be made to the scientific staff. No factor was 
more important in determining the past and present animal and human life of 
the plateau than climate. Geological and palaeontological data both indicated 
that there had been a succession of climatic cycles since Jurassic time, and 
possibly even before that remote period. Confirmation of this important induc- 
tion could best be furnished by palceobotany. Moreover, a collection of the 


recent vegetation on the plateau was of great importance, not only for itself 
but in connection with the zoological studies. 

Dr. John C. Merriam, Director of the Carnegie Institution, came to our 
assistance by offering to detail Dr. Ralph W. Chaney to the Expedition for the 
1925 season. We could not have had a happier choice. Although a palaeobot- 
antist by inclination, Doctor Chaney has a wide knowledge of geology, recent 
botany and zoology. His interest in all of these subjects greatly enhanced his 
value to the Expedition as a palaeobotanist. 

We also felt that the time had come for an archaeologist to join our party. 
While all of us had done what we could in searching for evidences of primitive 
hiiman occupation in the Gobi, the necessity for a trained man who could 
devote his whole time to this important science, was apparent. Mr. N. C. 
Nelson, who has had long experience in studying the primitive archaeology of 
Europe, was our unanimous choice. Mr. Nelson discovered surface indica- 
tions of early human occupation in Mongolia from almost our first camp 
throughout the entire traverse. We had covered much of this same region in 
our previous expeditions without making any archaeological discoveries, so 
that Nelson's work demonstrated how essential was a trained investigator. 

An expert topographer we all felt to be essential. Professor Morris had 
done this work on our first two expeditions, but this together with geology was 
too great a task for one man. Doctor Berkey suggested his friend. Major L. B. 
Roberts, U. S. Army Reserve, to handle the topography, and his appointment 
was strongly urged by the Chief Engineer of the United States Army. Major 
Roberts was responsible for the development of new methods of rapid map- 
ping and needed all his ingenuity to keep pace with the Expedition. For his 
assistants, I selected Lieut. F. B. Butler, U. S. A., and Lieut. H. O. Robinson, 
First Royal Lancashire Regiment (British) : young officers who also rendered 
much assistance in ways other than mapping. 

Doctor Berkey arranged to rejoin the Expedition for the 1925 season, as 
did Mr. Shackelford. We obtained the services of Mr. Norman Lovell (British) 
as Assistant Motor Transport Officer and of Dr. Harold Loucks of the Peking 
Union Medical College as Surgeon. The Expedition, as thus reorganized, 
included a foreign staff of fourteen men, and the sciences of palaeontology, 
geology, palasobotany, archaeology, zoology, topography and photography. In 
addition, Mr. Pope continued his work on herpetology and ichthyology in 


By the middle of Jime, 1924, I had completed all necessary arrangements, 
and sailed for China, reaching Peking on July 4. 

The 1925 Expedition was to explore a new route to Tsagan Nor and then 







3! o 

n 9 


'•^ - S*" '-■ 


continue westward along the northern slope of the Altai Mountains into 
untouched country. In order that we should not be subjected to delays in 
obtaining permits from the Mongolian Government, I decided to go to Urga 
to make the necessary arrangements in advance. 

I left Peking at the end of August, accompanied by Norman Lovell, 
F. A. Larsen and Gordon Verecker, Esquire, First Secretary of H. B. M. 
Legation, Peking. Mr. Verecker's object was merely a pleasure trip to Urga. 


It was a summer of unprecedented rains, and in August Kalgan was 
partly destroyed by a flood. Hundreds of thousands of tons of loess were 
brought down by the river which flows through the center of the city. The 
bridge was carried away and all the business sections flooded. Shop doors 
were literally sealed with mud. The Chinese take such calamities philosophi- 
cally, and they slowly dig themselves out, leaving ridges in the center of the 
streets, to be worn down by traffic as time goes on. A week of intelligent work 
would have cleared the entire business section, but no one thought of doing it 
after his own particular doorway had been opened. It was with the greatest 
difficulty that we forded the river at Kalgan with our car. 

The rain had also reduced the Wan Ch'uan Pass to a chaos of rocks and 
ruts. Only a motor of the strongest construction could have negotiated the 
Pass. The trip across Mongolia to Urga was pleasant and imeventful. The 
unusual rain had induced a wonderful growth in the grasslands, and even the 
Gobi was so clothed with green sagebrush that it hardly resembled the normal 
grim desert. 


We were required to submit to the search of our persons and effects by the 
insolent Buriat officials in Urga and to the still more annoying investigation 
by the Soviet Secret Police, even though I had notified the Government author- 
ities by telegraph of our arrival and we really were coming at their invitation. 

Fortunately, otir friend, Mr. T. A. Badmajapoff, was in Urga and he tinder- 
took to conduct negotiations on our behalf with the Premier and Ministers of 
Education and Foreign Affairs. It was a delicate business and, of course, was 
carried on without regard for time. A few weeks or months meant nothing in 
their lives, and the fact that I could not remain in Urga all winter, awaiting 
their pleasure, was simply my misfortune. I fotmd that a Scientific Committee 
had been formed, apparently for our benefit. Its object appeared to be to 
obstruct our scientific work as much as possible, for its members had conceived 
the idea that the fossils we were taking out of the covmtry were worth himdreds 
of thousands of dollars. Then the Secret Service took a hand and said that we 


could not be allowed to make maps or do any topographic work whatsoever. 
It was a pretty difficult situation, but, guided by the astute diplomat, Badma- 
japoff, I made considerable progress. At the end of three weeks the authori- 
ties had agreed to permit the Expedition to work in Outer Mongolia imder cer- 
tain conditions. These included the presentation to Urga of a duplicate set of 
all our collections, publications and maps. To house the material they would 
build a Museum. Moreover, we must take with us two Mongolian representa- 
tives, one from the Scientific Committee and the other from the Secret Police. 
These gentlemen were to act as the recording angels of our actions. I was to 
pay their expenses and salaries as well. The passports for ourselves, cars, 
camels and equipment cost three thousand dollars. It was agreed that they 
would be sent to me in Peking, for at least a month would be required for their 
preparation. It was exasperating beyond words, after all our previous work, to 
meet the same suspicions as in former years. The Buriat "advisers" pointed 
out that it was not reasonable to think that we would come from America, have 
a great caravan of camels and a fleet of motor cars, bring forty men and spend 
himdreds of thousands of dollars, just to dig up a few old bones. That was too 
ridiculous for any intelligent man to believe. Of course, we had an ulterior 
motive. Perhaps it was to search for oil or minerals; possibly we were spies; 
but certainly there was some dark reason for our presence in Mongolia. It 
was exactly the same type of argimient that we met in 1922. Fortunately, our 
previous record in the country was so good that even these new oflficials, who 
had been in Urga only a short time, were at last convinced that the Expedition 
would not be a serious menace to the safety of the State. 


One delightful incident of the Urga visit was the opportunity to meet 
General P. K. Kozloff. For years I had admired this famous Russian explorer 
who had been a companion of the great Prjevalski. I had had several letters 
from KozlofT and felt an intense desire to know him personally. He had made 
preparations for a two-year expedition into the southern Gobi in the region of 
Kara Khoto, imder the auspices of the Leningrad Academy of Sciences. The 
day before he was to leave Urga for the desert, a telegram was received from 
Moscow directing the authorities to prevent him from starting. General 
Kozloflf was completely at a loss to imderstand the reason for his detention 
and was unable to get any information concerning the situation. He had been 
a general under the Czar but had devoted his life to exploration and taken no 
part in politics. 

Kozloff, although nearly sixty-five years old, is a man of such energy 
that he could not long remain inactive. Being an enthusiastic ornithologist 
he began to make a collection of birds of the Urga region. One of his trips took 


him into the forest some sixty miles north of Urga, on the Kiachta road. He 
discovered a series of large mounds, overgrown with trees and underbrush, 
but of such a regular character that he suspected they coiild not be natural. 
After obtaining permission from the Urga authorities to investigate he sunk a 
shaft in the center of one of the largest mounds. About fifteen feet under- 
ground he found a rectangular chamber containing a coffin and other objects. 
These were subsequently identified as belonging to the T'ang Dynasty. A 
number of tombs were opened and the materials sent to Russia for study. It 
developed into such an important investigation that Kozloff postponed his 
proposed trip to the southern Gobi vmtil 1926. 

While I was in Urga he invited me to accompany him to the excavations 
and inspect his work. I went down the shaft into one of the larger tombs and 
spent a most interesting day examining the materials and surroimdings. Koz- 
loff 's enthusiasm at sixty-five was as great as though he had been a boy of 
twenty. I cotmt it as an exceptional privilege to have had the opportunity 
to meet this great explorer of the old school and to see him in action. 


Before I left Urga the Premier assured me that oiw various permits for the 
next summer's work would be issued in due course. Without the aid of Mr. 
T. A. Badmajapoff we should never have been able to overcome the suspi- 
cions and to reach an tmderstanding with the Urga authorities. We owe him a 
debt of gratitude not only for this splendid work on our behalf but also for his 
previous assistance to the Expeditions. He worked for us in the most unselfish 
way and, being so thoroughly respected by all factions in the Government, his 
championship of our cause was most valuable. He had been a member of one 
of Kozloff's earlier expeditions, as well as having accompanied us in 1922, 
and has a keen interest in exploration and scientific research. 


Some weeks after returning to Peking, our passports, motor car and camel 
permits arrived in the care of Mr. F. A. Larsen. I devoted myself during the 
winter to the preparations for the coming Expedition. Four thousand gallons 
of gasoline and one hundred gallons of motor oil were especially packed by the 
Standard Oil Company at the Tientsin office tinder Mr. Lovell's supervision. 
Two and one half tons of flour, a ton of rice, half a ton of sugar and other sup- 
plies made a mountainous heap on the laboratory floor. Lovell and I packed 
all the food and equipment and took it to Kalgan by train. 

I ordered Merin to bring in the camels, and on February 20, 1925, the 
caravan left for its long trip into the Gobi. I gave Merin instructions to go 
straight to Shabarakh Usu and await us there. He was to proceed by slow 


marches, resting the camels wherever he found good feed, but must arrive at 
the Flaming Cliffs before May i . Our base would thus be w^ell-advanced for the 
long westward exploration, and the camels fresh. While we were working at 
Shabarakh Usu, the caravan would push on rapidly along the northern base of 
the Altai Mountains. Armed with a sheaf of Urga permits for the camels, 
we anticipated no trouble for Merin in crossing the frontier into Outer Mon- 
golia. After the caravan left, Lovell and I occupied ourselves in preparing 
the outfit for each of the men, inspecting and sorting the motor equipment and 
getting everything in readiness for immediate departure after the staff arrived. 
By April 7, we were all assembled in Peking and ready to leave for the desert. 



Nearly all the existing maps of Mongolia are based on the Russian map, 
which is very tmreliable. Apparently much of it was prepared from native 
information, and this is proverbially bad. Mountains appear where there are 
no mountains. Many of the largest caravan trails are omitted, and important 
natural features are located incorrectly. In consultation with Major L. B. 
Roberts, we determined to make our survey more accvu^ate than any other that 
had been attempted previously on the central Asian plateau. Instead of rely- 
ing upon aneroid barometers for altitude. Major Roberts decided to carry his 
line of levels from the Kalgan railroad station, with a plane-table and Gurley 
telescopic alidade. Barometers were not used on the main route survey at any 
time during the season. 

It was by no means easy to carry out the initial and most important stage 
of the survey. Marshal Feng Yu-hsiang, who then was in charge of the Kal- 
gan region, had refused to permit another foreign geological party to make a 
survey of the Pass. There was no probability that we would fare better at his 
hands. In such cases, in China, I have found that the best plan is to go ahead 
until you are stopped. It is quite possible to make tinending trouble for one- 
self by being too conscientious in the observance of rules. Therefore, I 
instructed Major Roberts to say nothing to the authorities, but to do his job. 
If he got into trouble, I would guarantee to get him out of it. 

Major Roberts left Peking for Kalgan with Lieutenants Butler and Robin- 
son on April II, 1925. On the 13th, very early in the morning, they repaired 
to the railroad station, set up the plane-table and stadia rods, and began the 
survey. The few soldiers posted at the station were too lazy and ignorant to 
do more than glance at their operations, and did not report them to their 
officers. By the end of the day Roberts had carried his levels ten miles or more 
beyond the station and returned to Kalgan to sleep. The next day, with their 
camp gear piled in a cart which proceeded by road, the topographers reached 



Wan Ch'uan, and by April i6 had completed the line to the village of Miao 
T'an, on the plateau, where they were to await our arrival. They had not been 
molested by the soldiers, and had carried out their work according to schedule. 
The summit of the Pass was determined as 5007 feet above sea-level. 


On April 15 the rest of the staff left Peking for Kalgan, with the excep- 
tion of Shackelford and myself. Shackelford wished to make further tests 
with the radio set, upon which we were depending for time signals to check 
our chronometers. We went to Kalgan on April 16 and met Yoiing, who had 
been up to Chap Ser to bring down our three Mongols. He reported the Pass 
to be in such bad condition, due to recent rain, that I decided to wait a day 
for the road to dry. That evening Doctor Berkey developed a temperature of 
104°, and Dr. Loucks was greatly worried. He feared pneumonia. Rather 
than wait for Berkey 's recovery in Kalgan, where the staff would be idle, I 
decided to go on as far as Ula Usu and send back for Loucks and Berkey. At 
Ula Usu everyone could be occupied with effective work. Berkey was better 
next morning, April 18, and Dr. Loucks had hopes that he would be able to 
travel by the end of the week. Knowng Berkey's indomitable will and his 
remarkable constitution, I felt sure that he would be out of bed before then. 


At nine o'clock in the morning April 18, 1925, we left Kalgan, and found 
that the road had dried considerably. Moreover, Marshal Feng Yu-hsiang 
was planning to send out five or six motor cars to attack bandits who had been 
raiding on the plateau. In preparation, he had several himdred soldiers work- 
ing on the road, rolling out the rocks and filling in the deep cart ruts. 

When our cars started for the Pass they made a ver>' impressive specta- 
cle. The huge trucks, flying American flags, and piled high with baggage and 
men, looked like battleships under full steam. It would take a pretty strong 
band of brigands to worry us. Between the summit of the Pass and Miao T'an 
the going was so bad, due to recent rains, that one or another of the cars was 
mired every few miles. At about five o'clock a duststorm began and continued 
with such violence that we were only just able to reach the Chinese inn at 
Changpeh-hsien. During the last two miles we were almost blinded by the 
dust, and had to grope our way to the village. 


As usual, bandits had been active along the himdred miles of road from 
Kalgan northward to the end of Chinese cultivation. Several great caravan 
trails converge toward Kalgan, and the villages offer protection to the robbers. 


A week before we left, three American cars, loaded with eighty thousand dol- 
lars' worth of sable skins, had been robbed within fifty miles of Kalgan. Other 
automobiles had met a similar fate, almost every week. Before we started, the 
Foreign Commissioner at Kalgan had demanded that I sign a statement releas- 
ing the Chinese authorities from all responsibility for our personal safety. 
Otherwise, we woiild not be allowed to proceed. Although it was a bad pre- 
cedent to establish, I felt quite certain that we were much too strong a party 
for brigands to attack. As long as Chinese bandits can rob defenseless cara- 
vans they maintain a bold front, but they dislike exceedingly to be shot at 
themselves. Foreigners, as a rxile, shoot much too well, and they do not follow 
the Chinese custom of running at the first sign of danger. Twenty Chinese 
soldiers, be they bandits or otherwise, to one foreigner, is about the proper 
ratio for anything like a good fight. 


The next morning, Sunday, April 19, we were checked at the river five 
miles beyond Changpeh. The bridge had been carried away and while the 
water was then only three feet deep, the bank on both sides had been cut by 
carts into a mass of gluelike mud. Hundreds of camels, carts and ponies, 
besides two hopelessly mired cars, were waiting at the river. It seemed impos- 
sible to cross, but, by careful wading at the ford and testing every foot of the 
bottom, we managed to find a passable route. A huge Chinese car loaded with 
ammvmition was half sunk in the center of the stream. The soldiers were in 
despair, but we rigged a block and tackle and had it out for them in a half an 

Proceeding to Miao T'an, we repacked the cars and sent the topographers 
forward to carry their line up the road. In the meantime I took several of the 
men out to shoot geese, which were feeding in thousands on the green blades 
of wheat in the fields. The next morning, shortly after leaving Miao T'an, a 
Russian car reported fifteen bandits ahead, but they disappeared before we 
arrived. Roberts and his men carried their line forward with amazing speed, 
because while we were in the cultivated area they had houses and other objects 
for fixed "shooting" points. Later, on the plains, topographical work became 
more difficult and other methods were developed. 

We stopped for the night five miles south of Chap Ser, in the bend of a 
little stream which the Mongols called "Wolf Creek." Dry, yellow grass made 
a clean floor, and, in spite of the fact that it was our first camp, the tents 
were up in less than half an hour. After dinner we all gathered in the mess 
tent about an argul fire, for the temperature dropped to -i-20° F. 

The next morning, when we were all packed to leave, I discovered that my 
car would not start. Investigation disclosed a broken timing gear. The Dodge 


Brothers engineers had been so certain that this could not possibly happen, 
that we had not included any of the gears in our assortment of spare parts. 
There was nothing for it but to remain where we were while Young returned to 
Peking for a new gear. At the same time he could bring back Berkey and 
Loucks. While our camp was an excellent one in most respects, I was some- 
what worried because it was in the center of the bandit region. Most of the 
robberies had taken place within a few miles of Wolf Creek. Nevertheless, we 
had no choice but to remain. 


Young left Wolf Creek for Kalgan at 10:30 A. M. April 21, and I drove up 
the road to the topographers, who had gone ahead of the main party. I 
instructed them to proceed as far as our Mongol village, 120 miles from Kalgan, 
where we were to leave the main road. Then they were to return to camp. 
The next day some of the men went off to a considerable lake, six or seven 
miles to the east, to shoot ducks and geese. Nelson prospected the surrotmd- 
ing region for artifacts but fotmd nothing of interest. Morris was busy on 
geological problems. Early the next morning Shackelford and Butler received 
a message over the wireless from Young, which read, "Berkey well but weak. 
Leaving Kalgan Saturday, April 25." It was a bad start for the season, but I 
have found that a bad beginning often is followed by a good ending. 

The topographers kept themselves busy in making a detailed map, on the 
scale of I to 20,000, of the Hallong Ossu region where we were camped, showing 
the typical grassland topography. It extended as far as the Swedish Mongol 
Mission station in charge of Rev. Joel Eriksson. I knew the Mission people 
well, and later Mr. Eriksson became the Expedition's agent in Mongolia. He 
had had considerable training in medicine and surgery, although not a grad- 
uate doctor, and his medical work was a great boon to the Mongols of the 

Mr. Eriksson took Granger and Morris over to the fossil beds at Ertemte, 
which Doctor J. G. Andersson' had discovered in 1919-1920. They did not 
attempt to do any work there, as we had not yet been informed that Doctor 
Andersson had abandoned his investigations. For a few days the weather was 
beautiful, with warm, bright sun, little wind, and clear, cold nights. Toward 
the end of the week, gales began, which made the topographic work difficult 
and uncomfortable. Major Roberts and his men kept at it, in spite of the 
weather, and by April 26 had completed the area map. As we also obtained a 
splendid collection of the mammals of the region, the time was by no means 

' Andersson, J. G. 1923. "Essays on the Cenozoic of Northern China," Mem. Ceol. Survey of China, 
Series A, No. 3, p. 46. 

h' ' 





; .\ 

'\ , 


PS a 

o Si 

§ ^ 






U'^ tj; 


Shabarakh Usu. IO-25- 




It was interesting to see how Chinese cultivation was being pushed north- 
ward toward the edge of the desert. Every two or three years the Chinese 
open a new tract for settlement and, as soon as this is occupied by farmers, 
the border line of cultivation is advanced still farther. In the last seven years 
it has gone northward forty miles. The Mongols who occupy the frontier 
region are simply forced to retreat as the Chinese progress. The advance of 
cultivation is dependent, to a large extent, upon war and bandits. When 
political conditions are disturbed the rate of settlement is slowed down. It 
is much like the former frontier days in America, except that the Chinese 
farmers make no effort to defend themselves against the bandits. 


We discovered very soon why Marshal Feng Yu-hsiang, the so-called 
"Christian General," had objected to our going out to Mongolia. He knew 
that a clash with Chang "Tso-lin was inevitable in the not far distant future, 
and was preparing himself with arms and ammimition from Russia. Great 
quantities passed oixr camp daily in motor cars, and fifteen hundred camels 
were reported to be south of P'ang Kiang. Since Feng had categorically denied 
his Russian affiliations, he was not anxious to have foreigners on the road who 
would tell the truth. 


Although our camp at Wolf Creek was fortunately located so that most of 
the men could find something to do, we were delighted to be able to leave when 
Young arrived on Svinday, April 26, with Doctors Berkey and Loucks. They 
had encoimtered such a terrific sandstorm, at the top of the Pass, that they 
could not travel the previous day. Berkey was still weak, but had made a 
remarkable recovery under Doctor Loucks' careful nursing. 

We left at 7:30 the following morning, April 27, and turned off the road 
to the west at our village twenty miles from Chap Ser, which we named 
"Tserinville" after Tserin, one of our Mongols. Running through beautiful 
rolling grasslands toward the Sair Usu trail, we made good progress until the 


Then the topographers found that the country had become so flat that 
they could no longer use natiu-al objects for sights with the alidade. Major 
Roberts immediately put into practice a method which he had devised for such 
an emergency. In Peking, when discussing his plan of carrying line and levels 
from the Kalgan railroad station, he had met with the usual derision. He was 


told that in a flat country, where prominent natural objects were almost non- 
existent, it would be impossible to make a rapid survey. He met this difficulty 
by using the motor cars as stadia rods. The height of the hub, fender and 
windshield were known, and these points were sufficient, since the length of 
sight was from one to two miles. The distance was taken by speedometer, 
which had been carefully rated. The scale of the route maps was i : 200,000 
and the contour interv^al was 50 feet. 

Two other cars were detailed to assist the topographers and to act as 
foresight and backsight stations. While the backsight was being taken on one 
of these cars, the other would run on and pick up a point for foresight. The 
backsight having been completed, the rear car would be signaled, run forward, 
and while the foresight was being taken, pass the topographers and be ready in 
turn to establish another foresight. The average day's run with this system 
was fifty-eight miles and the maximum seventy-three and a half miles. The 
levels were carried by the table itself, and in many of the set-ups the topog- 
raphers had level readings. Usually it was possible to step the vertical differ- 
ence with the stadia hairs, and only rarely was recourse had to the Beaman arc. 

The speedometer on the topographers' car was frequently checked against 
courses from fifteen to twenty miles long, which had been laid out by stadia. 
The use of the speedometer in measiuing distance was more or less of an 
experiment, but the manner in which the traverse closed in on the control 
points fully justified the confidence in its use. Naturally, since the differences 
in elevation were frequently a function of the distance run, the question of the 
accuracy and suitability of the speedometer distance was one of double impor- 
tance. In certain instances where the nature of the ground was such as to 
cause an excessive error to arise from the use of the speedometer, the strip 
would be crossed by using the stadia. Sights as long as twenty-five miles were 
used in locating mountains and other features adjacent to the line. 

To one who has not worked with this or similar methods, the value of such 
a reconnaissance may seem questionable. The restdts of the season's work 
proved the system sound, however, and possible of wide development. In 
addition to the check afforded by closure on control points, there were oppor- 
tunities to intersect on distant, prominent mountain peaks, from various points. 
The invariable small triangle of error was a constant check on distance and 
orientation of the base-line. In addition to barometer comparisons, a check 
on line-levels was afforded by vertical angles read to these peaks from various 
stations along the route, when possible. Several times each day magnetic 
north was plotted on the route-field sheet, the solar chart being used to estab- 
lish the meridian. Establishing meridian in such a way was sufficiently accu- 
rate to be commensurate with the compass needle used. When the agonic line 
of the Eastern Hemisphere was crossed, much local attraction was detected and 


many reversals recorded, all of which made the topographers appreciate the 
solar chart as their most valuable aid. 

Whenever we were in camp for a day, latitude and longitude were deter- 
mined. Latitude was fixed by observation of Polaris off the meridian, the hour 
angle computed and correction applied. For time and longitude the equal- 
altitude method was used, both sun and stars being observed. This work was 
in Lieutenant Butler's hands. 

We carried with us three Hamilton chronometers; one was a No. 950 
(watch movement) and two were torpedo-boat type clocks. We had expected 
to check our clocks with time signals by radio, but when we reached Outer 
Mongolia the authorities woxild not allow us to use the radio set. Therefore, 
time, upon which the locations ultimately depended, was a weak factor. 
Traveling in motor cars and covering such a great distance in a day, the 
chronometers were subjected to frightful conditions. To offset this, certain 
stations were occupied for long periods at a time and the rating of the clocks 
could be determined over these periods. 

Likewise, on the run-back many of the original stations could be reoc- 
cupied after even a longer period of time, so that excellent field ratings were 
obtained. From the last of such observations to the final check-in and rating 
upon the return to Peking, a minimum of time elapsed, so that the over-all and 
intermediate rate has been established accurately. The error, as computed 
from the rates determined in the field, coincided in a gratifying manner with 
the compared error as found by radio check in Peking. The rate for the three- 
week period following our return made a smooth curve with the plotted field 
rates. In 1926 we were in constant touch by radio with the Naval Station at 
Cavite, Philippine Islands, and checked these same clocks against time signals 
nearly every night. At many of the 1925 stations, latitude and longitude 
observations were taken in 1928, and were found to check most satisfactorily. 
This, in brief, is the topographic method which Major Roberts developed and 
followed throughout the summer of 1925. We have no hesitation in presenting 
his maps as by far the most acciu-ate topographic work that has been done as 
yet on the central Asian plateau. 


Our camp the evening of Monday, April 27, was sixty miles from Wolf 
Creek, beside a well on the grassy floor of a dried lake bed. Shortly before 
reaching this place we saw a Mongol on a camel, who waved to us to stop. 
He had been sent by Merin to report that our caravan was being held by the 
border guards, about fifty miles from Ula Usu. They had been there a month, 
because the guards had found ammunition in our equipment. As a matter of 
fact, there were several boxes of shotgun shells in one of the cases, but the 


Urga authorities had assured me that their permit covered whatever our camels 
would carry, and that the caravan would be allowed to pass the frontier with- 
out examination. Since I knew from previous experience the type of insolent 
Buriat officials who are in charge of every border station, there was little doubt 
in my mind as to why our camels were being held. Every Chinese caravan is 
treated in the same way until the merchants pay enough "squeeze" to satisfy 
the greed of the Buriats. 

The detention of the caravan was a serious blow to all our plans. Instead 
of having our base at Shabarakh Usu, with rested camels awaiting our arrival, 
it had covered only half the distance with four hundred miles of the worst 
desert to cross. It meant that I should have to readjust the entire work of the 


We reached Ula Usu on Thursday, April 30, and found the place little 
changed since our last visit in 1923. It seemed to be a little more desolate, if 
possible, but that was all. Still the antelope did not find it so, for thousands 
of them were scattered on the plain all about us. We had passed herd after 
herd on our way from Wolf Creek, all of them the grassland species, Procapra 
gutturosa. I have never seen quite as many antelope at any other time as we 
found that year along our line of march. 

A lama came to camp and reported that the soldiers were keeping our 
caravan so closely guarded at the yamen that the camels had not been allowed 
to find proper pasturage; that all our boxes had been opened; that the roads 
were guarded by soldiers ; and that when we arrived I was to be taken to Urga 
and shot. A most interesting prospect! The next morning I took three cars 
and five men, Granger, Shackelford, Loucks, Young and Lovell, for the drive 
to the yamen. We were all heavily armed and determined to have the caravan 
with or without a fight. The Buriats could suit themselves as to which it 
would be. 

Fifty-two miles up the trail we fovmd five Mongol soldiers awaiting us. We 
unceremoniously bimdled one of them into the car as a guide to the yamen, 
and a short distance farther on discovered one of our own camel men. Twenty- 
five miles of bad cross-country travel brought us to the yamen. It consisted 
of two large yiirts surrounded by half a dozen tents. Not far away was our 
own caravan, with the American flag floating in the wind. Old Merin and his 
Mongols greeted us like joyous children. I found that considerable damage 
had been done to our supplies, but it was not as bad as we had feared. Half a 
dozen sacks of flour had been ripped open ; several tins in each box ptinctured 
and nailed cases pretty well ruined. 

As soon as we arrived, an insolent yoimg Buriat from the yamen swag- 


gered into the tent and said that I must prepare to leave for Urga in charge of 
soldiers at once, while the other men and camels remained under guard at the 
yamen ; that the head man would send for me later. 

"Tell yoiu- chief that we are ready to see him now" was the reply I sent 
back. Following closely on the messenger's heels, all six of us approached the 
yurt. The yoimg Buriat reappeared immediately, saying that the official wotild 
not receive us then. I asked Loucks and Shackelford to remain in the cars with 
rifles ready, and to use them if the Mongols tried to "start anything" outside. 
Then I lifted the door-fiap and stepped inside, followed by Granger, Young 
and Lovell, with Buckshot and two of otu" Mongols, Tserin and Aiochi. About 
twenty Mongols sat in a circle staring at us, fascinated. I said nothing for a 
few moments and then suddenly demanded, "Who is the head man?" A lama at 
the far end of the yurt, wearing a gorgeous yellow satin coat and sable-bordered 
hat, raised his hand. 

"How dare you ignore the passport of your government and hold our 

The lama replied in a low voice that he had found shotgun shells in two of 
the boxes. I told him that the permit covered everything the caravan carried; 
that he had opened our boxes without authority; that he had acted like a 
brigand, and that we intended to take him to Urga to answer to the govern- 
ment. At this point I brought my fist down with a bang on the sheet-iron stove. 
The lama, who had been nmning his beads faster and faster through his hands, 
snapped the string and crumpled his rosary into a yellow ball. Finally he 
managed to stammer that he wanted to pass the caravan, but that his soldier 
colleague would not do so because it contained ammunition, and also dan- 
gerous, seditious literature, in the form of Asia, World's Work, Outlook, Satur- 
day Evening Post and other magazines. Moreover, he had discovered a large 
box of "Eveready" flashlight batteries, which he thought were bombs, and 
two old Chinese bayonets used by us for digging in fossil beds. Altogether it 
was most dangerous to the peace of Mongolia! 

The lama and all the men in the yurt were so badly frightened that they 
were anxious only for us to leave with our caravan. I instructed Merin to get 
ready for an early start next morning, and spent the evening in taking a month's 
supply of food and gasoline from the boxes. These ignorant officials had made 
our work extremely diffictdt. Instead of going west directly upon our arrival 
at Shabarakh Usu, we shovdd have to remain there imtil the caravan arrived. 
I told Merin to travel fast, leaving the weakest camels by the trail. We must 
sacrifice the animals in order to make up as much as possible of the lost time. 
Merin thought that he could reach Shabarakh Usu in three weeks, and I gave 
him twenty -five days as a maximimi. 

We foimd the same difficulties whenever we came in contact with the fron- 


tier guards. The yamens are on all the main caravan trails, and mounted 
soldiers continually patrol the border. All of them are Bimats, and I know of 
no more insolent type of himian being than a Buriat in possession of a little 
authority. Without a show of force, we could get none of them to recognize 
the passports and permits which I had obtained from Urga at great expense 
and trouble. Every time we crossed the border we had to let the yamen 
officials understand that we would enforce our rights, with bullets if necessary. 
The government officers in Urga were politely regretful when I complained, 
but did nothing to remedy the situation. As a matter of fact, I think they 
make no attempt to stop the border yamens from oppressing the Chinese and 
other traders. 



The camels left the yamen on Saturday, May 2. The caravan made a 
beautiful pictiore as it moved into the morning mist which hung low above the 
desert. There were only one hundred and ten camels, as some had been 
released by the loads of gasoline left at Baron Sog-in-Sumu, near Ula Usu, and 
remained there for the summer. 

We returned to Ula Usu the same day, by a trail which runs from Baron 
Sog to Gatun Bologai, the spring and oasis that Young and I discovered in 
1923. It was considerably better traveling, as well as being more direct, than 
the Sair Usu road over which we had come. In camp the men awaited our 
rettim with the greatest interest, and there was general rejoicing that the 
camels were actually imder way. 

Roberts and his assistants had been working hard on a map of the Ula 
Usu fossil beds. Olsen foiuid a fine titanothere foot and Loucks discovered a 
complete front limb, but we decided to leave them imtil ovu" return. Berkey, 
Morris and Nelson drove to Jisu Honguer for ftirther study of a doubtful 
geological problem. 

Nelson fotmd a series of pictographs, pecked upon the rocks. These were 
easily recognizable as delineations of hiunan beings, camels, horses and cattle, 
as well as stags, antelopes, ibex and mountain sheep. I will describe these 
petroglyphs, and associated moimd structiu-es, more fully in a subsequent 
chapter. Near the camp at Ula Usu, Lovell picked up four iron arrow points, 
and others were discovered at various points along the route. Doubtless these 
were early Mongol. 

On May 5 we left Ula Usu, going to the obo three miles west of camp, 
and then across country southward to the trail on which we had returned the 
previous day. Granger, Chancy and Nelson assisted the topographers. The 
arrangement worked out well, for the frequent stops for sights enabled Chaney 
to collect plants along the way and gave Nelson an opportimity to search for 



artifacts. They both found many valuable specimens which would have been 
missed had they continued rapidly with the fleet. 


About two o'clock in the afternoon, when we were forty-three miles from 
Ula Usu, I was some distance ahead of the others and discovered a fine expo- 
sure of red sediments. It was too good to pass without further investigation. 
A series of five springs, named Baiying Bologai, had formed an intermittent 
rivulet about which sand-grouse swarmed in thousands. We made camp on a 
level plateau, with the stream below us. Shackelford and I immediately took 
posts near the water, with our guns, and proceeded to collect dinner for the 
camp. I never have had more sporting shooting. With the wind behind 
them, the birds came in like bullets. I made a record of fifteen birds with 
sixteen shells. We got forty-two in an hour. 

The topographers and others arrived at four o'clock in the afternoon, and 
after an early dinner all of us scattered over the badlands to prospect for fos- 
sils. Before long a number of bones of giant sauropod dinosaurs were dis- 
covered, thus identifying the strata as Lower Cretaceous. Unfortunately, the 
enclosing matrix was of coarse gravel and all the bones were badly broken. 
After a further investigation, Granger was convinced that anything we might 
find would be in the same condition and not worth removing. It was evidently 
a deposit formed by swift-flowing water. Nevertheless, the discovery was 
important as locating another large area of Cretaceous strata. 

Lovell found an owl's nest containing nine white eggs in a rock hole. The 
disgorged pellets were formed almost exclusively of sand-grouse bones and 
feathers. The night was clear and cold. From Ula Usu to Baiying Bologai 
the elevations varied between thirty-seven and forty-two hundred feet. 


The following day was cloudy and cold and for the first time the 
topographers were forced to resort to the compass for obtaining their 
plane-table set-up. Leaving them to follow, we drove on twenty-eight 
miles past the yamen. With waving coats the officials tried to induce us 
to go over to the yurts but I saw no necessity to inconvenience ourselves 
since they already had seen our credentials. My feelings toward the inmates 
of that particular yamen were such that were I to write them I would need 
asbestos paper! 

We had tiffin at the base of a bluff where Shackelford discovered a deposit 
of paper-shales containing plant and invertebrate remains. Although they 
were rather badly preserved I thought that Doctor Chaney might wish to make 
a careful study of the locality. I left a note on a stick in the middle of the trail 











saying that if Chaney desired we would camp for a day at Gattm Bologai. 
When the topographers reached the spot they delayed long enough for Chaney 
to make a representative collection of the shales. It was identical with the 
Lower Cretaceous, Ondai Sair formation at Uskuk, but the plant specimens 
were not as well preserved as in the latter deposit. 

The topographers and geologists arrived at camp at Gatim Bologai at 
about seven o'clock, having surveyed fifty-eight miles during the day. The 
run had been over elevations varying from thirty-two to thirty-nine himdred 
feet. When we left in the morning, sand-grouse were coming in thousands to 
drink at the spring. The air literally was black with them as they swept down 
to the water in dense clouds. 

Retracing our way six miles, we found a trail which the Mongols assured 
us ran straight to Artsa Bogdo. The terrain was hard gravel, and except for 
one bad sand- wash in a field of "nigger-heads" the going was excellent. Thirty- 
five miles from the road-fork we met oiu- caravan in camp. The camels were 
doing fairly well, although considerably weakened by their close confinement 
at the yamen, where they had not been able to obtain proper feed. Merin 
promised to reach the Flaming Cliffs in a fortnight. Near their camp we saw 
a well-marked trail which Merin said ran directly to the temple at Sain Noin 
and the Orkhon River. 

Not far from where we met the caravan, I saw one of the small, rare 
McQueen's bustard, Chlamydotis undulata macqiieenii, a species which we had 
first noticed near Iren Dabasu. 


In a dry river-bed we found a dozen isolated elm trees. It was always a 
great cause for rejoicing when we saw trees in the Gobi. One does not realize 
how greatly one has missed them, and even though they were old and marred 
by their battle for life still they delighted our eyes. 

As I have remarked, the water-table in the desert usually is less than 
twenty-five feet below the surface, so that it is not remarkable that mature 
trees are able to reach water with their roots. The significant point is that 
there are no young trees or saplings to be found at any of the tree-bearing 
localities in the desert. Since there is no evidence of a change in the soil, the 
logical inference is that at the present time there is considerably less moisture 
than in the past when the trees got their start in life. The aridity is so great 
that saplings are not able to get a foothold. Moreover, only the most vigorous 
of the old stock have been able to maintain themselves to the present time, and 
their unhealthy condition prophesies an early demise. This is only one evi- 
dence of the minor fluctuations which have occurred within the larger climatic 



We camped at three o'clock in a dry, sandy river-bed where there were 
more ekns. The elevation was only three thousand feet. It was difficult 
pitching the tents because of a high wind, which pulled the iron pegs out of 
the soft ground almost as fast as they were driven in. Close to the trail was a 
well of excellent water, named Baiying Goshigo. Unlike any other wells that 
we have seen in Mongolia, this had sides built up six or seven feet above the 
surface. The camp was fifty-seven miles from Gatun Bologai. We had 
learned that the topographers under ordinary circumstances covdd not do 
more than sixty miles of mapping in a single day and even that brought them 
into camp thoroughly exhausted. They never had an idle moment when the 
Expedition was on the march. 

Granger and I recognized this well as not far from the spot where Johnson 
had killed a mountain sheep in 1923. We rejoiced to know that the Mongol 
information was correct and that the trail would lead us to Shabarakh Usu. 
During the night Merin arrived with the camels. 


In order to facilitate work while traveling, the Expedition was divided 
into two imits. Young and Lovell with the two Fulton trucks and Shackelford 
and I in the touring car were in the advance party. We had with us all the 
camp -gear and servants and ran ahead as rapidly as possible, breaking trail 
for the topographers. Wherever there was a particularly bad spot we left a 
note with directions as to how to negotiate it. In the second unit were the 
topographers with Chaney and Nelson occupying one car and Granger and 
assistants in another. 

The geologists, Berkey and Morris, who usually remained behind with the 
topographers, were always kept busy studying outcrops and keeping their 
route section intact. Sometimes they had to nin several miles off the trail to 
investigate an interesting exposure. Nevertheless, every member of each unit 
knew where the others were and kept in constant touch in case of trouble. If 
the geologists had gone a considerable distance off the trail and did not return 
in a reasonable time, either Granger or Chaney and Nelson went to investigate. 
It often happened that a car got mired or in such a position that it could not be 
extricated without assistance. It never was safe to lose contact with any mem- 
ber of the Expedition. 

With the advance party I drove a mile or two in front of the others. My 
job was to pick out a trail, find a way around or through bad spots and thus save 
the heavier trucks from getting into avoidable difficulties. Also I shot enough 
gazelle or sand-grouse along the way to provide meat for camp. When we 
had gone about sixty miles we pitched the tents at the nearest water. Usually 


we woiold arrive about three or four o'clock in the afternoon. The topog- 
raphers' unit seldom got in before six o'clock and often later. They found 
camp made and dinner ready. 

At the end of the 1925 Expedition we discontinued the use of the Fulton 
trucks and had an entire fleet of Dodge Brothers cars. The trucks were useful 
because they were very strong and carried an enormous amount of baggage 
and men, yet their weight continually got us into difficulties. Often where 
the lighter cars would run over the surface, the heavier trucks broke through 
the crust and we had distressing delays and hard work. 


Friday, May 8, was a hard day for all of us. We left the camp by the 
trees rather late because the sand in the river-bed was so soft that we had to 
take the cars to hard ground and carry the equipment to them. The first 
twenty-five miles was fine going but then Young's truck broke through the 
crust in the middle of a great field of "nigger-heads" and sank to the axles. 
Getting it out was a slow process. As usual, the only possible way was to col- 
lect stones, or anything solid, and put them londer the jack by the wheels. As 
the jack pushed them down others were added. Eventually the foimdation 
was strong enough to bear the weight of the car as the jack lifted it up. Some- 
times one of the other cars could give assistance in pulling out a mired car, 
but as a rule such attempts were useless imtil there was a firm foundation under 
the wheels. Hardly had Young's truck been released from its bed of muck 
when we drove into soft sand. Every car had to be pushed across half a dozen 
bad spots. 


When we camped after fifty-two miles, at a well called Khimdelungi Usu, 
all the men were exhausted. It was time for some artificial stimulation, and I 
opened one of the twelve bottles of whiskey which had been brought for special 
celebrations. None of the foiuteen men covdd have very much to drink from a 
single bottle, but the mental effect was quite as good as though there had been 
twice the amoimt. As a result we had a merry dinner in spite of tired bodies. 
A terrific wind all day long had added to the discomforts and nerve strain, 
particularly of the topographers. It was extremely difficult to keep the plane- 
table steady and get accurate readings. Nevertheless, they had siirveyed 
fifty-two miles and found that the elevations for the day's run were between 
twenty-nine hundred and thirty-six hundred feet. During the day we crossed 
a river-bed lined with ekns. They were old and gnarled but made a most pic- 
turesque avenue as they followed the winding stream course for more than a 



As usual after a stretch of bad going, we found good traveling the next 
day and ran up and down a series of gentle slopes and across smooth plains, 
gradually coming up to a height of five thousand feet. The country was so open 
and the weather so fine that I knew the topographers could get long sights. 
We expected a record and they made it by siu-veying nearly seventy-two miles. 

The desert was extremely arid aU through this region — like a clean-swept 
floor of gray gravel with hardly a trace of vegetation. We did not pass a well 
for fifty-four miles. While the terrain was excellent for the motors, I wondered 
what the camels would find in the way of feed on the hard, bare floor. Merin 
told me later that his caravan had a distressing time in this region. The only 
way was to cross the rocky desert as quickly as possible even though the lack 
of food and water took its toll from the weakened animals. Of course, there 
were no Mongols in this inhospitable part of the desert. 

Jicm OLA 

Our camp that night was beside a well on the slopes of a long, low, ragged 
ridge of sediments called Jichi Ola. The going had been so good that oiir 
advance party did the seventy-two-mile run in four and a half hours, arriving 
at one o'clock in the afternoon, but the topographers did not reach camp until 
seven in the evening. The surrounding desert was fairly well covered with 
vegetation and abundant grass grew in the narrow valleys which thrust into 
the mountain. Three yurts were pitched a short distance from the well and we 
found their inhabitants most friendly. For gas tins and other small things 
we obtained a supply of dry argul. "Buckshot" made the best bargain by 
getting a Mongol to bring up two pails of water from the well and accept, with 
the greatest delight, two nails in payment. Five or six miles to the north in 
a field of "nigger-heads" there were half a dozen yurts and a considerable herd 
of cattle, ponies and sheep. Near the well we saw a fair-sized wild sheep's 
head and during a walk on the mountain LoveU, Loucks and Tserin saw three 
rams and five ewes. On the return journey we spent several days here shooting 
and had a most interesting time. 


The next morning we came in sight of the Gurbun Saikhan shortly after 
leaving Jichi Ola. On either side of the trail were low, isolated and very rough 
mountain ridges, the easternmost end of what farther west develops into the 
Altai Mountain chain. The mirage was so well developed that the topog- 
raphers could not take sights of more than half a mile. Our day's nm amovmted 
to only forty-three miles, but imder the adverse conditions even that distance 
gave the topographers more than they could comfortably handle. 



On the way to Shabarakh Usu next morning, we had a surprising exhibi- 
tion of the ability of a gazelle to run when badly wounded. I shot a young 
buck, completely severing the hind leg at the knee. The animal continued to 
run at a speed of twenty-five miles an hour for five miles. The going was so 
bad that the car could not reach a greater speed than that and we were not able 
to get near enough for a second shot. Two wolves also gave us an exciting 
chase but we lost them in a series of deep ravines. 


Late in the afternoon we reached the eastern end of the great Shabarakh 
Usu basin. My car was in the lead and I had worked up to the opposite slope 
before the others picked their way into the chasm. They looked like tiny 
black ants on the surface of a great red wall. One could not conceive of them as 
automobiles. Evening shadows already had transformed the vast pink basin 
into an unreal place. Like a fairy city it is ever-changing. In the fiat light of 
midday the strange forms shrink and lose their shape; but when the sun is 
low the Flaming Cliffs assume a deeper red, and a wild mysterious beauty lies 
with the purple shadows in every canyon. 

There had been little change since we left in 1923. The tracks of our 
motor cars were filled with sand but still distinct; the old camp site on the 
basin's rim was marked by a heap of discarded stone blocks, each containing 
an incomplete skuU of a dinosaiir which had lived there in the far dim past. 

Our tents were pitched on the basin floor near the well. I loved the spot, 
for I had only to raise my eyes to see the sctdptured ramparts of the Flaming 
Cliffs shimmering in the Gobi mirage. A few hitndred yards to the north was 
an area of old dead sand-dunes now dotted with a "forest" of so-called tama- 
risks, Salicornia herbacea. These are stimted desert trees which reaUy are not 
tamarisks at all. They are under fifteen feet in height, yet Doctor Chaney 
found by sections that many of them were more than two hundred years old. 



We were all very tired from the continued traveling, and I announced 
that the day after our arrival, May 12, would be one of rest. We could shave, 
bathe, cut hair, read and write to our hearts' content. But few of the men 
could loaf for a full day in such an interesting spot. Shackelford was busy 
with his camera, Nelson wandered off into the sand-dvmes and Granger gave 
the fossU beds "the once over." He reported that the two seasons of rain and 
frost and blasting gales had removed a good bit of surface sediment and that 
there were excellent chances of new palaeontological discoveries. 


Shackelford had been looking about in the "tamarisks" during the day 
and reported finding many artifacts. In the evening I walked over to the 
dvmes and into a shallow valley, the floor of which was thickly strewn with 
such specimens. The next morning Nelson and I went out immediately after 
breakfast, followed by Berkey, Morris and Loucks. We found an area of dead 
dunes which had been "anchored" by the small trees. Sculptured red bluffs 
marked the entrance to shallow valleys floored with soft sandstone where the 
wind had swept away the loose sediment. 

On the clean surface, flakes of red jasper, chert, chalcedony and other 
white stones were scattered like new-fallen snow. Pointed cones, neatly 
shaped where flakes had been stripped off, tiny rounded scrapers, delicately 
worked drills and a few arrowheads gave Nelson the first indications of the type 
of culture with which he had to deal. We held a consultation. "Where did 
the artifacts come from? Could they have been washed down from the sur- 
face?" These were the first questions to be answered. We must find flints 
actually in the rocks and bones to date the deposit geologically. 

Shortly after our constaltation I discovered a bit of eggshell of the giant 
ostrich Struthiolithiis . A few yards to the left Morris found another fragment 
of shell drilled with a neat round hole. Nelson said it was one of the beads in a 
necklace. We were in a fever of excitement for the trail was getting hot. 
Nelson, the most conservative of conservatives, was skipping about from place 
to place like a boy of sixteen. At last Berkey foimd a spot where half a dozen 
chipped flints were deeply embedded in the sandstone floor. We made much 
of him for his discovery only to find that it had already been marked by 
Shackelford the previous day. Before noon we had discovered a dozen 
such spots and were satisfied that some, at least, of the artifacts had weathered 
out of the lowest level and had not washed down from the surface of the dunes. 

An tmlooked-for complication entered when we began to discover frag- 
ments of pottery. It was primitive enough, to be sure, but a people who used 
such crude stone implements had no business to be making pottery! The 
problem became more interesting and more complicated every hour. 


I have never seen the advantages of correlated work more clearly dis- 
played than in solving this human problem. The geologists, palaeontologists, 
topographers and botanist aU assisted the archaeologist. Without such a com- 
bination of expert knowledge available on the spot, it would have been impossible 
to settle many of the puzzling questions presented by this deposit. The sub- 
ject became so interesting that it was difficult to keep all the men from hunting 
artifacts. Doctor Loucks was one of the most enthusiastic workers. In com- 
pany with Doctor Berkey, he discovered a vast workshop where flint chips 
were scattered over the surface in tens of thousands. They took four of our 
Mongols to the spot one morning and returned with about fifteen thousand 
flakes. Nelson was busy for days sorting the pile and selecting such as were 
valuable for specimens. 

The second day's work revealed dark spots in the lowest layers of the soft 
red sandstone. Evidently these were ancient fire sites. When they were dug 
through in cross sections, layers of ash containing charcoal, flints and burned 
bones were revealed. Very soon we fovmd square bits of dinosaur and ostrich 
eggshells embedded in the sandstone. It was then that we realized that these 
Dune Dwellers were the original discoverers of the dinosavir eggs. About that 
time Doctor Loucks found quantities of the ostrich eggshell on the surface of 
the peneplane. Evidently these primitive people picked up both the ostrich 
and dinosaur eggshells at the Flaming Cliffs two miles away and brought them 
to their workshops here in the dunes. A few mammal bones were found 
embedded in the flint -bearing strata but they were so poorly preserved that 
identification was impossible. 


On the second day after a good many flints had been found in situ and 
Nelson had come to the conclusion that we were dealing with a Mesolithic 
culture, Doctor Chaney played a joke which was most amusing. He found a 
bit of rusted iron saw blade and planted it neatly in the flint-bearing layer. 
Doctor Berkey was the one who discovered it first. This caused utter con- 
sternation. It completely upset all our theories and gave us a bad hour. But 
while we were sitting disconsolately about the spot racking our brains to 
accoimt for its presence, Nelson strolled up and produced the other part of the 
blade which he had found near camp ! 

We determined to get even with Chaney. He was an enthusiastic collec- 
tor of bird's eggs and spent every leisure moment blowing and labeling them. 
Shackelford and I got two weU-matched hen's eggs and had the cook boil 
them hard. Then they were beautifully stained in potassium permanganate. 
I found a bush near the sand-dunes where the grotmd was splashed with bird- 
droppings, scooped out a hoUow depression and "set" the eggs. A pair of 


demoiselle cranes lived near the spot and I told Chaney that probably there 
was a nest in the vicinity. He never had seen a crane's nest or eggs so that the 
rest was easy. When I returned to camp and announced the discovery he was 
all excited. Four of us piled into a car and ran down to the spot. He was so 
delighted that I almost relented. Then I remembered the saw and hardened 
my heart. After Chaney had photographed the "nest" from three angles and 
made a close-up with the portrait lens, we went back to camp. Word had been 
passed around and eight or ten men gathered to see the denouement. First 
attempts at blowing were not successful and after a serious discussion as to the 
best method of preservation he decided to remove the embryo through a hole 
in one side. I never will forget the expression on his face when he discovered 
that they were hard-boiled ! With a roar he hurled one at Mac Young and the 
other at me, but we already had a good start across the desert. 

Practical jokes on an expedition may lead to bad feeling, but Chaney was 
a good sportsman. He knew that he deserved all he got after the episode of 
the saw. Still, he assured me that I headed his list of preferred funerals and 
that sooner or later I would "get mine." I did, too. It was when we were 
eating the last of the twelve hundred eggs that we had brought into the field. 
An innocent-looking, but previously prepared boiled egg was served me at 
breakfast and when I broke the shell a flood of pink water soaked everything 
on my plate. Then all debts were paid. 


While every day was bringing to light fresh information regarding the new 
human culture, the other work of the expedition progressed just as satisfac- 
torily.. The day after our arrival Buckshot and Liu each discovered fine 
Protoceratops skulls and the topographers had begim a detailed map of the 
basin. Chaney explored a bluff to the north of the sand-dunes where he found 
a deposit of shale containing poorly preserved plant remains and some wood. 
In a basin to the east. Granger also found wood which had been converted into 
charcoal by burning, probably in the course of an ancient forest fire started by 
lightning. The trees were of fair size, the largest two feet in diameter. Doctor 
Chaney is convinced that during the Cretaceous the climate of this region was 
too dry for the preservation of many plants. 


Olsen was particularly anxious to find more dinosaur eggs, and shortly 
after our arrival he discovered a beautifiil set of five in a most vmusual way. 
He was prospecting in the gully where he had discovered the first eggs in 1923. 
Not thirty feet from the site of the original nest, he saw a bit of shell fragment 
in the loose sand ; a few yards farther up the slope was a larger piece — then no 

^ F*^^v' ^^^^^- ■ ■■ ■■■- '^V . 


This set is in the Field Museum of Natural History, Chicago. 



more. Crawling on hands and knees he went over every inch of ground, but 
there was not a trace which could lead him to expect the presence of more eggs. 
Impatiently, he drove his pick into a cracked rock, overturning a chimk 
weighing fifty pounds. Adhering to the imder-side were five dinosaur eggs, 
three of them unbroken. The fourth was cracked and the end of the fifth fitted 
the fragments which had led him to the nest. The discovery was ptire accident 
because Olsen does not often waste time and energy in turning over rocks when 
he cannot see fossils. These eggs were much smaller than those we had fotind 
in 1923 and evidently were laid by a different type of dinosaur. They are now 
in the Field Museum of Natural History, Chicago, Illinois. 

Another accidental discovery of dinosaur eggs was even more remarkable. 
Norman Lovell was interested in getting young birds for pets. One day he saw 
a kite's nest just under the edge of the cliff that terminates the great peneplane 
which sweeps down from the Gurban Saikhan and stops at the basin. After 
several vmsuccessful attempts to climb the face of the cliff, he gave it up and 
approached the nest from the plain "to see what he could see." Crawling on 
his hands and knees to the very edge, he lay flat on his stomach trying to peer 
into the nest, when his hand struck something sharp. It was the knife-like edge 
of a broken dinosaur eggshell ! The upper parts were gone but the remains of 
eighteen eggs were in their original position firmly embedded in the rock. 
Perhaps in another few months of weathering this section of the basin rim 
wovild have broken away and the eggs smashed to bits on the rocks below. 
There was nothing but pure luck in the discovery because the only eggs in 
LoveU's mind were those of the kite which he expected to see in the nest 

It was a delicate and dangerous operation to remove the eggs. A high 
wind blew the entire time, and Walter Granger had to lie at ftill length to avoid 
being swept over the brink. He took out the block of stone containing the 
eggs and it was sent to the Museum entire. Although the exposed surfaces of 
the eggs are weathered and broken, the lower halves are intact and they make a 
superb exhibit. The eighteen eggs in the block stand on end in a rough double 
circle. Doubtless the nest originally contained more, for there was evidence 
that others had broken out as the edge of the cliff crumbled away. East of 
the Flaming Cliffs, Chaney picked up seven hundred and fifty fragments of 
dinosaur eggshell in one spot. 


Even the topographers did their bit for pateontology, for while mapping 
the sand-dunes Robinson found many bits of StriUhiolithus shell embedded 
with worked flints in the lower sandstone layer. We were ciirious to know 
why these ostrich egg fragments should occur in such ntimbers in a single 


locality near the top of the peneplane. I went out with Berkey and Loucks, 
who had discovered the spot. We found what Berkey is confident was an 
ancient basin of redeposited Pleistocene sediments which probably made an 
excellent nesting site for the giant ostriches. 


On May ii and i6 there were such furious windstorms that it was quite 
impossible to leave camp. We could neither work, read nor write — nothing in 
fact but find whatever spot offered the maximum of shelter and sit there with 
faces wrapped in wet cloths. As usual after a really bad sandstorm, the fol- 
lowing day dawned without a breath of wind and with crystal -clear air. The 
sand and gravel being heavy do not remain suspended, and the wind seems to 
"wash" the atmosphere of all haze and fine particles of dust. The staff took 
up their interrupted operations with eagerness, and during the next four days 
the camp was deserted except at meal times. 


I had received word that the Government ofiicials insisted that I come to 
Urga. New regulations had been passed which required discussion if we were 
to continue our explorations. It was extremely annoying to have to leave the 
Expedition when we were in the midst of such interesting work, but there was 
no alternative. I decided to go up with McKenzie Young on May 24, and 
finish the disagreeable business as soon as possible. From Shabarakh TJsu we 
estimated the distance to Urga at about three himdred miles. The Mongols 
said that a small trail ran northward and that eventually it joined a well- 
marked caravan road which led to "the capital. Even without trails we felt 
sure that we could find the way easily enough, as we got exact compass direc- 
tions from Major Roberts. We decided to take only one car and Tserin, our 
usual Mongol companion. 


The twenty-third was a beautiful summer's day. The Flaming Cliffs 
whirled and danced in the heat-waves reflected from the basin floor. Beside the 
tent a huge black raven drowsed upon a rock with beak half open, and even 
the spotted lizards were too sleepily content to snap at the tiny sand-flies which 
crawled incautiously beneath their pointed snouts. It was so warm that as 
Yoimg and I prepared for our trip to Urga we thought only of the lightest 
clothing. But when we left next morning a raw wind was blowing from the 
north. Summer had departed in a night and it kept on going every hour. By 
noon the car could barely make headway in the teeth of an icy gale which 
whipped sand and gravel against our faces like bursting shrapnel. We have 


had too much experience with MongoHan weather to start on any trip without 
fur coats, but even in these our teeth were chattering. 

At seven o'clock in the evening when we saw a lone yurt beside the trail, 
both of us had had more than enough. The only inmates were a yoiuig lama 
and a wrinkled old hag, aged seventy, with a baby of four or five years. But 
they made us welcome and piled dry argul upon the fire until our niunbed 
bodies had begun to warm. In half an hovir the mother of the baby arrived — 
a strapping young Mongol with a frame like a football tackle. She had been 
out all day searching for a flock of sheep while the man toasted himself at the 
fire. When more argul was needed he never dreamed of moving but told the 
old grandmother to bring it from the pile outside. Tserin pitched our tent 
close to the yurt with the aid of the lama. Cold antelope meat, a tin of sweet 
corn, coffee and biscuits made otir dinner. We were too tired to be himgry, 
but the steaming coffee was better than the most wonderful wine. The circu- 
lar felt-covered yurt was like all others. On a red chest opposite the door stood 
a Buddhist picture and a few sacred family treasures. At one side was a plat- 
form bed raised four inches off the ground; the man slept on it, of comse, while 
the women had two rolls of felt nearby. On the other side of the ytut was a 
wooden rack holding bowls of curdled milk; close beside it two calves and 
half a dozen baby lambs and goats tied to a rope. The yixrt reeked with min- 
gled odors of mutton fat, rancid butter, unclean Mongol and living goat, but 
it was warm. 

Of course otir tent had been pitched with its back to the gale but during 
the night the wind made a complete shift. When I awoke in the morning I 
had the strange feeling of a heavy weight pressing me down. Opening my eyes 
I could see only white. Then I slowly began to realize that I was buried in a 
snowdrift. I managed to sit upright and saw only a great white pile where 
McKenzie Yoiing ought to be. At my call, strange heavings took place and 
McKenzie sat up looking wildly about at the tmfamiliar siuroundings. The 
tent was packed solidly with a drift which in the back reached right up to the 
ridge pole. All our clothes were buried deeply somewhere in the snow, and to 
put the cold, damp garments on our warm bodies was anjrthing but pleasant. 
The front of the car had been covered with a canvas hood. When McKenzie 
started the engine a peculiar muffled roar came from the interior. He found 
it was as solidly packed as though we had shoveled snow inside. 

It was eleven o'clock before we finally got away from the friendly yurt and 
began to fight otir way toward Urga. That day was a constant battle. Often 
when plowing through the white blanket on a smooth plain, the car wovdd 
suddenly drop into a ravine packed six feet deep with snow. It was a terrible, 
sickening feeling when everything seemed to go out from imder us. Then 
there was nothing for us to do but to dig around the car till it reached the 


ground, jack up the wheels and pave a road with stones until we could get out. 
Poor McKenzie had to bear the brunt of all the work, for my right shoulder 
was still virtually helpless from a crash in the New Year's steeplechase at 
Tientsin, when I had torn my collar-bone away from all its fastenings. I was 
of use only to pick up stones or drive. 

With the greatest difficulty we had worked our way across two ranges of 
high hills but stuck badly near sunset, half-way to the crest of a mountain pass. 
Two hours of heart-breaking work got us out of the ravines and we drove up 
to a Mongol yurt in the curve of the lower hills. The warmth and shelter of the 
tiny felt house were glorious even though we shared it with goats, calves, men, 
women and children, who watched our every move with fascinated eyes. Next 
morning we awoke to gorgeous sunshine, but the trail was deep in snow and 
we had to work far up the grassy slopes, circling and twisting to avoid the drifts. 
The tires slipped on the wet grass, and the car would move forward only 
where we built a path of stones. After five hours of dangerous and strenuous 
work, we were only three miles from our night's camp. But our troubles were 
at an end when we gained the svimmit of the pass and drove down the steep 
slope into a beautiful valley. There we got back to the trail, which became 
broader as we approached Urga. 

In the snow-filled ravines lay the bodies of horses and cattle that had been 
caught in the white death-traps. Some of the animals were too weak to move 
and gazed at us with pitiful eyes. I longed to end their sufferings with a bullet, 
but it would only have meant trouble with the Mongols. That blizzard took a 
frightfvd toll of life all across the northern grasslands, for it came so suddenly 
that the natives were vmprepared. 


From the crest of a hill Urga lay below us like a beautiful jewel set in the 
green valley of the Tola River. We could see the sparkling roofs and golden 
cupolas of the Living Buddha's palace nestling in the silver poplars at the base 
of Bogdo Ola, "God's Mountain." Towering above the city was the great 
temple surrounded by the cubicles of ten thousand lamas. Peaceful enough 
it looked in the spring sunshine, but I knew that it was a city of suspicion, 
and one not to be entered without due thought of how one was to get 

We had been especially invited to come to Urga by the governmental 
Scientific Committee. One object of our visit was to impack and install for 
them a large collection of fossils which the American Museum had presented. 
It included a replica of the giant Baluchitherium skull and of the famous eggs. 
In spite of arrangements which were supposed to have been made on our 
behalf, we were subjected to merciless searching of oiir persons and belongings. 









and every scrap of writing and all books were sent to the Secret Service office 
to be perused by censors. 


It was almost dark when we escaped from the last yamen and reached the 
house of Mr. Percy Marshall, Manager of the International Export Company's 
Urga branch. A warm welcome awaited us from Mrs. Marshall, although her 
husband was away. Oscar Mamen, then in the employ of the Mongolian 
Cooperative Company, arrived from Kalgan only a short time after we reached 
the house. Mr. Johannsen, a Danish trader, from Uliassutai, was another 
guest. The Marshalls' house was the one bright spot in Urga. It was sub- 
stantially built of logs, comfortably furnished, and Mrs. Marshall had given 
that touch of feminine charm which can come only from a woman of refinement. 
John T. McCutcheon, the famous cartoonist of the Chicago Tribune, and Mr. 
Barney Goodspeed with their wives had left for Kalgan the day before. We 
had hoped to see them, but they could wait no longer and started just in time 
to meet the blizzard which we encountered. Fortunately for them there was 
much less snow in the east and they got through with only two days' delay and 
comparatively little discomfort. 


The morning after our arrival we discovered that all the arrangements 
which I had made in 1924 with much labor and expense had been nullified by 
"new regulations" passed diuing the previous month for our especial benefit. 
"Laws" of this sort were mushroom growths which were rapidly ruining trade 
with Urga. When a caravan of cigarettes reached the outer barrier at Mai- 
mai-cheng and before it had passed the first inspection new regulations might 
be promulgated raising the duty on cigarettes fifty per cent. This actually 
happened in dozens of instances with various kinds of goods. As a result the 
merchants never coiold tell whether they were to face a profit or loss on any 
shipment. In 1924 I had spent about three thousand dollars and nearly a 
month of time in making arrangements for the Expedition. All of it was lost 
and it was necessary to begin at the very beginning. As usual, Mr. Badma- 
japoff was of enormous help. 


The Mongol Government presented impossible demands. We were to be 
required to bring all our collections to Urga, where the Scientific Committee 
was to take whatever it wished. No maps of any kind could be made and no 
geological work done. Moreover, the Mongol Government was to designate 
two students who were to be sent to America and educated at Harvard Uni- 


versity entirely at the Expedition's expense. It required days of talk and nego- 
tiations before a reasonable agreement could be reached. In the meantime 
we were under constant espionage. Tricks of various sorts were resorted to, 
whereby the authorities hoped that we might do, or reveal, something which 
would show that our scientific work was merely a camouflage for some sinister 
design. I must say, however, that I can hardly imagine more clumsy attempts 
of surveillance than the Secret Service instigated in our case. The first day 
the poor spy who was detailed to follow us was nearly exhausted, because on 
foot he could not keep pace with our rapid trips in the motor car. Therefore I 
invited him to occupy it with us, much to his enjoyment. He turned up at 
nine o'clock every morning to reUeve the agent detailed to watch the house at 
night. Two hours of continued questioning at the Secret Service office netted 
them nothing, and with the greatest reluctance they had to admit that we 
appeared not to be as potentially dangerous to the safety of the covmtry as 
they had hoped. 


One very pleasant break in our stay at Urga was again to meet General 
P. K. Kozloff and his charming wife. They had been in Peking and returned 
with Mr. Marshall. The evening of their arrival I walked over to the tiny house 
in which they were staying and found the General and his wife in the midst of 
skinning a dozen large hawks. These they had collected on the way to Urga 
and had started their preparation before resting or having dinner. Such 
enthusiasm in a man, then sixty-six years old, was a splendid inspiration. 
Kozloff was preparing for his expedition to Kara Khoto which had first been 
delayed by orders from Moscow and then by his subsequent discovery of 
Tang Dynasty tombs in the forests north of Urga. We made plans for our two 
expeditions to meet at the Altai Moimtains near Shabarakh Usu, but they did 
not materialize. Kozloff was again delayed and did not leave Urga until late 
in the autumn. 


On Jime 4, the last docimient had been signed, the last permit received and 
the last dollar paid. I felt that I should go mad if I had to make another visit 
to the yamens. The agreement which we finally signed with the Scientific 
Committee was fair and provided that certain duplicates of our collections 
should be returned to Urga. 

Nevertheless, we were required to take two Secret Sei-vice agents with us 
on the Expedition. One was openly to report our movements; the other mas- 
queraded tmder the guise of a member of the Scientific Committee. AU this 
after we had already conducted two seasons of exploration in Mongolia, and 


the entire world knew what we had accompHshed ! It was annoying to have 
the two Buriats with us, as I felt sure that through ignorance and misunder- 
standing they might cause us trouble. My fears were realized. 


When we concluded the negotiations at Urga, it was five o'clock in the 
afternoon, but there was still time to get through the barriers and well out upon 
the open plains before dark. On leaving Urga we were joined by the two 
Buriats who rode with Tserin on the top of the loaded car. One was Dalai 
Badmajapoff , whom we took at the request of Mr. T. Badmajapoff, as a guest 
of the Expedition. He spoke excellent English and was a very nice boy. We 
were glad to have him. The other was the representative of the Scientific 
Committee, John Dimschikoff. He was a "professor" in one of the schools. 
The subjects which he "taught" included every known branch of science. 
Since he was only twenty-four years old, there were still a few things which he 
had not learned, but he would not admit as much. 

Dimschikoff had also been commissioned by the Secret Service as a con- 
fidential spy to ascertain from our daily conversation what were our real 
motives in coming to Mongolia. Since he could speak only a limited amount 
of English, he understood just about half of what we said. He evidently 
thought that he would gain much credit if he "discovered" something startling 
regarding us, and sent in reports of the most amazing character. One of them 
was to the effect that Mr. Gordon Vereker, First Secretary of H. B. M. Lega- 
tion, Peking, was a member of the Expedition in disguise. Mr. Vereker had 
accompanied me to Urga the previous autumn merely for the experience of the 
trip. Dimschikoff reported that he had learned from our conversation that 
America and England were planning to combine for the purpose of annexing 
Mongolia. There was much else in the same vein. The Secret Service in Urga 
were only too ready to believe Dimschikoff's reports, but Badmajapoff per- 
suaded them to wait. When Dalai and John returned, the former repudiated 
all that the latter had written. However, except for the fortunate circimi- 
stance that by that time we had passed beyond the frontier of Outer Mongolia, 
we should have been arrested and taken to Urga. 

The Secret Service sent another man who joined us later when we were at 
Tsagan Nor. He was a very decent fellow, and his reports with Dalai's con- 
firmation completely discredited the wretched Dimschikoff. Evidently the 
latter was sent away from Urga to his home in Siberia because of the reports 
he had manufactured concerning the Expedition. Two years later he had 
sufficient influence to be allowed to return to Urga and was sent to Germany 
with Dalai. He stole the money for their expenses and poisoned Dalai. He 
then reported that the latter had taken the money and committed suicide when 


he, Dimschikoflf, discovered the theft. Our experience with Dimschikoff is an 
excellent example of what unfortunate results may happen when an expedition 
is forced to take such men into the field. As a rule they do not have the 
interests of the expedition at heart; on the contrary they feel that they can 
win recognition for themselves by creating trouble. If the Leader were allowed 
to choose the men who were to accompany him, it might be a very different 



Our return to Shabarakh Usu was very different from the northward 
trip when we fought snow and cold at every mile. Spring had come again and 
brilliant sunshine flooded the grassy hills. We passed a dozen lakelets dotted 
with breeding water-fowl. From the shore of one rose the sheer wall of a 
granite mountain, warped and folded into grotesque shapes. Late in the after- 
noon of the second day the grass began to thin ; turf was replaced by fine gravel, 
and we realized that we were entering the northern edge of the desert. Our 
camp that night was only eighty miles from the Flaming Cliffs. We hoped to 
reach Shabarakh Usu by ten-thirty in the morning, for the vast peneplane was 
like a floor, but blow-outs of two of our tires halted us tmder a ridge less than 
two miles away. With fleld-glasses we could see the blue tents steaming in 
wavering lines along the heat-soaked plains or dancing in a graceful mirage 
above the basin floor. We gave our emergency signal of three rapid shots, but 
the men in camp could not discover us in the fluid sunlight. 

Leaving McKenzie with the car, I walked in toward the tents through the 
grove of stvinted "tamarisks" on the dead sand-dunes. I was amused to watch 
the reaction in camp at my arrival. One of the Chinese servants first discovered 
me. Shading his eyes, he peered intently, then called another boy. At last 
they were convinced that it really was I. Then Lovell burst from his tent, fol- 
lowed by Shackelford and Robinson. Anxiously they ran to meet me, thinking 
that some accident had happened, but I shouted that all was well and to send 
out a car with extra tires for Mac. 

It was good to be back in camp. A fortnight had passed since we left, and 
it was like reading an absorbing novel to hear the story of what had happened 
during my absence. Most of the men had gone for a day's trip to the Gurbun 
Saikhan, but I heard an outline of events from Shackelford before he read his 
mail. More dinosaur eggs had been found. The excavation of Lovell's nest 
had been completed, and Doctor Loucks had discovered another "setting" of 



very small thin-shelled eggs quite unlike any of the others. Nelson's work on 
the new human cvilture was ended. They had located a dozen other artifact 
sites and had cut through many fireplaces but no hviman bones were discovered. 
Berkey and Morris had come to the conclusion that the flint-bearing deposit 
in the dead sand-dunes represented a new geological formation, which they 
named the Shabarakh formation. 


The way in which the basin at Shabarakh Usu was made is so interestingly 
and clearly stated by Doctor Berkey that I quote his remarks in full : 

"At Shabarakh there is an ancient river valley eroded to a depth of over 
400 feet below the general level of the Gobi planation siu-face. It could not 
have been made under present climatic conditions. There must have been 
enough of a stream in some former time to do this work. Now there is no 
stream at all. Closer inspection shows that the sediments, now well exposed 
by present-day erosion, which is accomplished chiefly by the wind, are not the 
same as those forming the sides of the valley into which the river trench is cut. 
Whereas the strata of the valley side carry dinosaiir bones of Cretaceous age, 
these sediments in the valley bottom carry only fragments of modem forms and 
hiunan artifacts. The differences in lithologic character of the two lots of sedi- 
ments throw further light on the matter. The valley bottom material exhibits 
evidence of deposition under rain wash and torrential conditions alternating 
with spells of wind action and with development of dune sands as time went on. 
The lower beds are therefore dominantly torrential sands, alternating with 
silts laid down in small ponds of standing water, whereas the higher beds show 
more and more the features of windblown material. 

"Topographically the valley bottom shows certain abnormalities also. 
The cross profile of the valley is not regular and the longitudinal profile and 
gradient are likewise more irregular than one should expect for a valley eroded 
in uniform sediments. These irregiilarities suggest obstruction of the valley 
by deposits of some kind. 

"Clearly the whole series of observations coupled with those made in the 
adjacent Altai Mountains, which there is not space to discuss, support the fol- 
lowing explanation. The original Shabarakli valley was eroded to its maximiun 
depth under much more hvunid conditions than at present. There was enough- 
rainfall to maintain an active river, and to support vegetation on the valley 
slopes which were thus protected from excessive rain wash and under these 
conditions maintained a smooth outline. But there came a permanent change 
of climate that has prevailed with only minor variations ever since that time. 
The region became more and more arid and there was insiifiicient water to 
maintain enough of a stream to continue valley erosion. Increasing aridity 


resulted in destruction of vegetation so that the valley slopes became bare 
and thus lost their protection against attack by the infrequent rains and the 
more constant winds. The occasional rainstorms therefore caused excessive 
wash of soil from the steep valley sides, and this debris collected in the valley 
bottom, whence it could not be removed by the reduced stream. In the 
intervals between storms the winds whipped the dried sands about and piled 
them into dunes and shifted them along the valley bottom, thus adding to the 
irregularity of distribution. This combination of processes must have con- 
tinued for a long time for the valley bottom became more and more heavily 
covered and obstructed with these deposits. 

"The increasing aridity resulted first in the loss of permanent stream 
flow, but for a time there were still struggling streamlets and transient pools 
dammed in by the obstructing wash from the valley sides. In these pools 
the finer sediments were collected, but tdtimately even these ponds were dried 
up, so that there was water only occasionally after a storm or during the rainy 
season. Finally, after layer upon layer of such deposits had been accumulated, 
the climate became so much more arid that even the rain wash from the valley 
sides was not siifficient to balance the destructive work of the wind, which with 
perhaps increasing vigor continued its attack on the lower sand. So effective 
and dominant had wind erosion become by this time that the finer sands were 
carried entirely out of the valley and the attack continued on the vinderlying 
deposits. Thus erosion has continued with short periods of reversed con- 
ditions down to the present day. This deposition history, followed by the 
reversal of operations from deposition to erosion, is fully recorded in the bare 
and much dissected remnants of strata where one can see every structural 

"These are the deposits in which and on the surface of which the artifacts 
registering the occupation of man have been found at Shabarakh Usu. A 
primitive race of men lived in the valley bottom along the course of the ancient 
stream. As the conditions changed on the last great cyclic swing toward a 
warmer and more arid climate, they gathered in the valley bottom, the only 
place where water cotild be obtained, and in this way they held out for a long 
time against the increasing privations of the encroaching desert. Repeatedly 
the sites of their hearths and workshops were flooded and overstrewn with 
wash from the hillsides due to a sudden storm, and the products of their work- 
manship were covered to become part of the permanent sedimentary deposits. 
Such experiences must have continued for some thousands of years until the 
accumulation filled the whole valley bottom to a depth of probably a himdred 

"The deposits formed during this time therefore carry evidences of the 
culture of a long period of time. In this connection it is especially important 


to note that the artifacts of the lower beds are of a much more primitive type 
than are those in the upper layers. Not by any means all of the original 
deposits are exposed to inspection ; the oldest and lowest beds are still covered 
and the highest ones have been destroyed; but a cross-section of the inter- 
mediate members is open to inspection. As a consequence, of course, the ear- 
liest stages of the htunan history of the locality cannot be read here because 
the records are still bviried ; but the later steps are marked by finds still in place 
in successively higher and higher layers of deposits; and the very latest 
stages, which must have been recorded in overlying beds that are now 
destroyed, are mixed together in the jumble of residuary debris left from the 
selective erosion work of the wind. At such places these reUcs of the handi- 
work of prehistoric man may be found literally by the thousand."^ 


Because of the fact that these primitive people inhabited the sand-dunes 
of the desert, Mr. Nelson named them the "Dune Dwellers" and says: 

"///. Mesolithic. This stratum alone yields a combination of traits which, 
so far as known, may be termed distinctively Gobian. Moreover, its normal 
occurrences in old, dead and indurated sand deposits, named by Messrs. 
Berkey and Morris, the Shabarakh formation, is so striking and distinctive 
that we venture to designate it the Shabarakh Cidture. The stratigraphic 
position of this Shabarakh Culture makes it positively pre-Neolithic; and the 
chipped stone (mostly red jasper) artifacts by which alone it is known, in sev- 
eral specific details conform closely to the Azilian flint industry of Western 
Europe. The inventory is as follows: 

1. Hammerstones, mostly roundish adaptations. Scarce. 

2. Cores, or nuclei, slender, crude, multi-faceted, angular and spherical. 


3. Cores, or nuclei, slender, oblong, cylindrical to conical in outline with 

sometimes a sharp projection on one side adaptable for cutting 
ptirposes. Numerous. 

4. Flakes, various forms, broad angular or subtriangular, with no retouch 

or indication of use; serviceable as knives, etc. Derived from Core 
No. 2. Very numerous. 

5. Flakes, long slender, prismatic, and often very delicate, with little or 

no retouch or sign of use; suitable as drills, gravers, knives, etc. 
Derived from Core No. 3 by pressiu^e process. Very numerous. 

' Berkey, Chas. P., and Nelson, N. C. 1926. "Geology and Prehistoric Archaeology of the Gobi 
Desert," Amer. Mus. Novitates, No. 222, pp. 7-8; Preliminary Reports, Central Asiatic Expeditions, Amer. 
Mus. Nat. Hist., II, No. 67, pp. 1-16, 1930. 

" ■ 


' V . ' 

r — ■ : — 1 — 1^— 







■ * 


■ ■' J 






V . i 

' r 








1 ' 

i. ■ 
1, ' 



,. :^:._/ . 

• ■ 





6. Perforators, produced by trimming or secondary chipping of flakes 

from either of the above groups. Rare. 

7. Retouched flakes, various forms, of imcertain purpose but of use as 

knives or scrapers. Moderately numerous. 

8. End-scrapers, small, mostly thumbnail size, round to oblong, some 

double-pointed, distinctly Azilian. Very ntimerous. 

9. Disk beads of Struthiolithus eggshell (occasionally dinosaur eggshell) 

in all stages of preparation, from rough ar jjtilar fragments to drilled 
and ornamented examples. Finished forms scarce. 

"/F. Neolithic. This stage is an outgrowth, with modifications and 
additions, of the preceding Mesolithic. The two agree also, for the most part, 
in mode of occurrence ; but the geographic range of the later Neolithic horizon 
would seem to be considerably greater. The outstanding traits include the fol- 

1. Hammerstones, ordinary roundish. Scarce. 

2. Grinding slabs of stone, thin, rubbed surface. Scarce 

3. Rubbing stones, flat, oval outline. Scarce. 

4. Mortars, globiilar. Scarce and even doubtfiol. 

5. Celts or axes, flaked surfaces with slight amount of polish near cutting 

edge. Scarce. 

6. Adzes or gouges (perhaps scrapers with curving edge) , chipped siorf ace. 


7. Cores — as imder III, 2. Numerous. 

8. Cores — as under III, 3. Moderately ntunerous. 

9. Flakes, ordinary — as under III, 4. Numerous. 

ID. Flakes, prismatic — as under III, 5. Moderately numerous. 

11. Perforators — as under III, 6. Moderately numerous. 

12. Worked or retouched flakes — as vmder III, 7. Moderately numerous. 

13. End-scrapers — as under III, 8. Rather scarce. 

14. Side-scrapers or choppers, elongated and disshaped. Moderately 


15. Blanks or roughly flaked but unfinished implements. Moderately 


16. Spear points or knife-blades (mostly represented by fragments) of 

ordinary Neolithic form and finish, i. e., chipped in Solutrean style on 
both surfaces; more or less lanceolate, with straight, convex, or 
stemmed butt ends. Moderately numerous. 

17. Arrow points of small, often delicate, sub-triangular outline, with 

straight, concave, convex, or stemmed butt ends. Rather scarce. 


1 8. Potsherds, color gray to brick red ; surface plain, string marked, incised, 

or stamped with geometric patterns, some laid on or modeled decora- 
tions. Moderately plentiful. 

19. Hearths, with the usual accompaniments of ashes, broken stone, bone, 


The quantity of later material which had washed down upon the floors of 
the valleys and was mixed with the flint implements eroding out of the lowest 
level, presented, at first sight, a most puzzling problem. It was only after 
careful study by Nelson, with assistance from almost every member of the 
Expedition staff, that the true situation was delineated. Doctor Loucks was 
indefatigable in searching hidden comers of the sand-dunes and the surrovmd- 
ing country. Chancy was almost as busy, and Shackelford during the intervals 
of photographing discovered flints in the most unlikely places. 

The discussions at night about the camp-fire of tamarisk logs were extraor- 
dinarily interesting and stimulating. Stones of various kinds were baked in 
the fire to test their manner of breaking under the heat for comparison with 
those found in the ancient hearth sites. After the surrounding region had been 
thoroughly explored it became evident that nearly all of the material which 
the Dune Dwellers used for their artifacts and hearths had been brought there 
from some foreign locality. But where their source of supply had been located 
we were unable to discover until later in the season. 

Since the story of Dune Dweller life in Mongolia was much more com- 
pletely pictured by discoveries made during the season of 1928, I will reserve 
a fuller discussion of it for a later chapter. 


At the eastern end of the dvines, Loucks found a cave containing a 
Mongol's cache. A complete yurt, floor skins, clothing and household utensils 
had been neatly packed away, and the door blocked up with a stone. It was an 
excellent commentary on the owner's confidence in the honesty of his country- 
men to leave such valuable material where it could easily be found by a traveler. 
I imagine, however, that the custom of the country made it quite safe, since I 
have frequently fovmd similar deposits in other places. 


During my absence in Urga, Doctor Chaney had discovered few fossil 
plants. It became increasingly evident that during the Cretaceous this region 
had been too arid for the preservation of such material. Nevertheless, he kept 
his presses full of the living flora, which gave dozens of forms new to his col- 

' op. cit., pp. 12-14. 


lection. Plants of the pea family dominated all the rest, presumably because 
they are particiilarly adapted to arid conditions. Yet there was an abundance 
of geraniums, buttercups, anemones and many new grasses. 


One of the letters which I had brought from Urga for Walter Granger was 
from Dr. W. D. Matthew. Matthew was one of the least excitable men I have 
known but he was really stirred when he wrote that letter. He said that a tiny 
skull in a nodule of sandstone discovered in 1923 and labeled by Granger "an 
unidentified reptile" was, in reality, one of the oldest known mammals. It 
had been found in the Cretaceous Djadochta formation which yielded the 
Protoceratops and dinosaur eggs. In a hundred years of science only one sktill 
of a mammal from the Age of Reptiles ever had been discovered. That single 
skull, named Tritylodon, from the Triassic of South Africa, is in the British 
Museum and was one of the world's greatest palaeontological treasures. But it 
belongs to the group known as the MidtitiihercnJata, which died out in the 
Eocene, and had no very direct relationship to living mammals. 

In his letter Doctor Matthew wrote: "Do your utmost to get some other 
skulls." Granger and I discussed it for half an hour, then he said: "Well, I 
guess that's an order. I'd better get busy." He walked out to the base of the 
Flaming Cliffs and an hotir later was back with another mammal skull! It was 
a similar experience to that of Professor Osbom at Irdin Manha when he told 
me that he was going to find the tooth of a Coryphodon and two minutes later 
picked up the second one ever discovered in Asia. Such things do not sound 
possible, I will admit, but they did happen. 

Granger's new skull was in a sandstone concretion but appeared to be 
well-preserved, as indeed it so proved. We had to leave for the west the next 
day, but when we returned in August, Granger and Olsen, "Buckshot" and 
Liu did some very intensive searching. It was close and trying work, for the 
skulls were in little nodules of sandstone that had broken out as the cliffs weath- 
ered away. There are literally millions of such concretions on the basin floor 
and it was simply a matter of examining as many as possible during the day. 
When one has inspected a thousand or more in the scorching sim with no 
result, the job loses its interest and becomes decidedly discouraging. Never- 
theless, Granger and his assistants stuck at it day after day. At the end of a 
week they had a total of seven skulls. Moreover, all but one had associated 
lower jaws, and several concretions contained skeletal material. It was pos- 
sibly the most valuable seven days of work in the whole history of paleontology 
up to date. 

Those skulls were the most precious of all the remarkable specimens that 
we obtained in Mongolia. In the field, Granger carried thera in his suitcase. 


I took them to New York and they were hardly out of my sight on the long 
journey from Peking to New York. With a good deal of relief I presented 
them to Doctor Matthew at the American Museum on November 9, 1925. I 
told him that this was the direct result of his letter. He had asked for "the 
goods" and Granger and Olsen had "delivered" them. Within a few hours of 
my arrival at the Museum, Albert Thomson began the preparation of the 
skulls. It had to be done imder a microscope and the hard rock particles picked 
off one by one with tiny needle -pointed instruments. It was such tense, nervous 
work that frequent rests were necessary. Still, by the beginning of the New 
Year, Thomson had finished the preparation and the skulls were ready for 
study by Drs. W. K. Gregory and G. G. Simpson, 1926. 

From Shabarakh Usu we obtained eight partial skulls with associated 
portions of lower jaws, one skull without jaws, a fragment of a maxilla and 
part of a mandible ; the remains of eleven individuals in all. These have been 
assigned to five genera and six species as follows: 

Djadochtatherium matthewi Simpson 



Deltatheridium pretrituherctdare Gregory and Simpson 

DeUatheroides cretacicus Gregory and Simpson 

Hyotheridium dobsoni Gregory and Simpson 


Zalambdalestes lechei Gregory and Simpson 

Zalambdalestes grangeri Simpson 


Regarding these archaic mammals, Professor Osbom' has remarked: 
"There is little doubt that the extinction of the large terrestrial and 
aquatic reptiles which survived to the very close of the Cretaceous prepared 
the way for the evolution of the mammals. Nature began afresh with the small 
unspeciaHzed members of the warm-blooded quadrupedal class to slowly build 
up out of the mammal stock the great animals which were again to dominate 
land and sea. One of the most dramatic moments in the life history of the 
world is the destruction of the reptilian dynasties which occurred with appar- 
ent suddenness at the close of the Cretaceous, the very last chapter in the Age 
of Reptiles. 

"We have no conception as to what world-wide cause occurred, whether 

' Osbom, Henry Fairfield. 1910. "The Age of Mammals," pp. 97-98. Macmillan, New York. 

'^ ".-'vIS 



-*'"' ■ • 




* ■ 







■» ' % 



there was a sudden or a gradual change of conditions at the close of the Creta- 
ceous ; we can only observe that the world-wide effect was the same : the giant 
reptiles both of sea and land disappeared." 

Dr. W. K. Gregory, who with Dr. G. G. Simpson, 1926, studied the speci- 
mens which we collected, says that these first primitive mammals were tiny 
things hardly larger than a rat, with fiorry bodies and pointed snouts. They 
scampered about tmder the feet of the dinosaurs and may possibly have had 
some significance in the extinction of the reptiles by eating their eggs. Or it 
may have been that they were better adapted to withstand changes in temper- 
ature, or had superior brains and a better method of moving about. But at 
any rate their progeny eventually became the dominating creatures of the 

Doctor Gregory asserts that the skulls we found have proved to be verita- 
ble missing links in the story of mammalian evolution. They show that even 
at the close of the Age of Reptiles, mammals as a class had separated into 
marsupials and placentals, as well as an older side branch called multituber- 
culates; also that the placental stock had already split up into various distinct 
families. "Hence it is not surprising," he writes, "that even in the five Creta- 
ceous placental mammal skulls found in one locality in Mongolia we have 
representatives of four genera and two distinct families. The more insectivo- 
rous-like forms in some ways suggest the tenrecs or centeloid insectivores of 
Madagascar, but both their skull characters and their dentition are definitely 
less specialized than in their modem relatives. The members of the second 
family are distinctly more carnivorous in the general form of the skull and 
dentition and they combine characters of the carnivorous-insectivorous marsu- 
pials with others seen among the earliest creodonts or primitive placental 
carnivores of the Eocene Age of North America. In both families the brain- 
case is distinctly smaller than in modem insectivores or marsupials of the 
same size. 

"Huxley, Henry Fairfield Osbom and other palaeontologists had already 
inferred that the remote ancestors of the higher or placental mammals in 
Cretaceous times were small forms of insectivorous-carnivorous habit, but 
prior to the Mongolian find the Cretaceous forerunners of the swarming placen- 
tal mammals of early Eocene times were practically imknown except by a few 
scattered fragments from the Upper Cretaceous of North America of very 
doubtful affinities."' 

Doctor Gregory further remarks that the Mongolian skiolls confirm the 
above view of Huxley and Osbom and that they lend support to a prophesied 
early stage in the evolution of the famous "tritubercular" type of molar tooth. 

' Gregory, William K. 1927. "Mongolian Mammals of the 'Age of Reptiles,' " Scientific Monthly, 
XXIV, pp. 225-231. 


Doctor Simpson writes, in prefacing a discussion of these specimens: 
"Not only are these remains by far the most complete ever discovered in 
the Mesozoic, but they also occupy a very strategic position in time and in space 
which makes close scrutiny of the relationships essential. In time they occur 
in the Cretaceous when, according to theories formed before their discovery 
and based largely on early Tertiary mammals, the differentiation of the 
placental orders should be in progress and not yet far advanced. In space they 
occur in central Asia in or near the region which a number of students, espe- 
cially Osbom and Matthew, have considered as an important center of radia- 
tion, and probably the very one from whence came the groups of mammals 
which appear to have entered North America and Europe suddenly at the 
beginning of the Tertiary and which must have been imdergoing an important 
deployment diu-ing upper Cretaceous time. The Mongolian Cretaceous 
insectivores are thus actual representatives so long hoped for but so little 
expected, of a group hitherto hypothetical and known only by its presumed 

In concluding his paper Doctor Simpson says: "The structure of the del- 
tatheridiids agrees with their position in time between the pre-placental, pre- 
marsupial pantotheres and the close but distinct array of placental orders in 
the early Tertiary and with their position in space near the center of the land 
masses later dominated by placentals in suggesting that they, of all known 
mammals, stand closest to the common point of divergence of many or all 
placental mammals. In the skull and dentition they come very near to showing 
all the features which the most competent students of Palaeocene and early 
Tertiary mammals have beUeved would characterize such a central group 
when found. "^ 

' Simpson, George Gaylord. 1928. "AiEnities of the Mongolian Cretaceous Insectivores," Amer. Mus. 
Novitates, No. 330, p. i; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., II, No. 85, 
pp. i-li, 1930. 

' Op. cit., p. 9. 



During the time that Young and I were in Urga in 1925 the caravan 
had arrived at Shabarakh Usu, and left after depositing supplies of food and 
gasoline. Granger sent the camels on to Tsagan Nor, for they were traveling 
slowly. The long march of four hundred miles across an almost foodless desert 
had drawn heavily upon their slender reserve strength. If the soldiers had not 
detained them at the frontier, they woidd have been at Shabarakh Usu a 
month before us, rested and well-fed. 

The men were only awaiting my retiun before starting for Tsagan Nor. 
We left June 8. Although expecting an easy trip we found much soft terrain 
in the valley bottoms. When we passed over it in 1922, it had been baked 
hard and easily supported the cars ; now they broke through the thin crust in a 
dozen spots and we had hard digging and pushing. Still there were long 
stretches of good going and we could not complain. Twenty-three miles west 
of Shabarakh Usu we saw two wild asses. This location up to that time was 
the easternmost record for the species. 

The caravan was camped at a tiny temple forty-five miles from the Flam- 
ing Cliffs. Several of the camels had died and the others badly needed rest 
and food. We left eighty cases of gasoline and all our specimens to await our 
return in care of the friendly lamas. Nelson reported great quantities of Dime 
Dweller artifacts along the trail and Chaney was enthusiastic over new plants. 

Artsa Bogdo was deserted by the Mongols. All otir friends were gone and 
not a yurt could be seen at the base of the mountains. We found the Mongols 
forty miles beyond, in the bottom of a beautiful valley watered by dozens of 
threadlike streamlets which twisted and turned across the meadow as they 
descended from the eastern end of Baga Bogdo. It was beautiful to look at 
but difficult for the cars to cross. We had to build dozens of turf bridges to 
negotiate the tiny streams. Our little lama hunter greatly longed to accept 
our invitation to accompany us westward, but he woiild first have to obtain 



permission from the Buriat supervisor of the district, who was camped some 
miles away, and we had no time to wait. Our friends told us that they were 
very unhappy because so many restrictions had been put upon them by the 
Urga authorities that they were no longer a free people. They could not sell a 
sheep or goat without reporting all the facts; they were taxed exorbitantly and 
could not leave their circumscribed localities without permission. To the 
Mongols, who are as independent in nature as the eagle, such conditions must 
be almost insufferable. 


On June lo, 1925, we sighted Tsagan Nor. What a disappointment! 
The beautiful lake of which we had talked so much to our new men was only 
a quarter of the size it had been in 1922. Wide stretches of evil-smelling mud 
encircled the water; the brilliant green vegetation had given place to a margin 
of dull yellow grass. When the topographers arrived, Nelson looked at it dis- 
gustedly and remarked: "Tsagan Nor! It's little and it stinks!" Neverthe- 
less, on the opposite side the beautiful yellow sand-dimes remained and the 
majestic peak of Baga Bogdo stood white and still in the brilliant simlight. 

We pitched the tents at the eastern end of the lake on one of the ancient 
shorelines which gave a fairly level shelf in the steeply sloping gravel bank. At 
our old camp site of 1922, the water was more than a quarter of a mile from the 
former beach. It soon became evident that bathing in the lake was not 
desirable. The mud was ankle-deep and as far as we could wade the water did 
not reach our knees. Three years ago it had been very slightly brackish; now 
it was strongly saline. 

After our first disappointment had abated we began to enjoy the camp. 
It certainly is the most beautiful spot that I have seen in all Mongolia. The 
mountain with the vast alluvial fans sweeping out from the black canyons has 
an indescribable mystery and beauty. In the soft colors of sunset the lake is 
changed to green and blue, making a perfect mirror for the snow-crowned 
stmimit of Baga Bogdo and the yellow mass of sculptured sand-dunes. After 
dinner we usually sat in the twihght smoking and talking of the day's work. 
Sometimes I climbed to the edge of the gravel plain to sit alone in the desert 
stillness, to observe lake and camp spread below. The subdued murmurings 
of sleeping water-fowl or the faint long-drawn howl of a lone wolf only intensified 
the silence. One by one the candle glows in the tents would disappear and the 
camp become a vast, dim shape upon the white rim of the lake. 


An old Mongol, who had lived all of his seventy-three years near the lake, 
told us that when we went there in 1922 the water was near its highest level. 













• ^ * V 


He said that forty and fifty years ago, Tsagan Nor was dry. He predicted that 
the water would disappear entirely that year, as indeed it did. There had not 
been siiflficient rain to fill the little river, Tatsin Gol, which gave it life. The 
springs under the lake likewise had been cut off because the successive dry 
seasons, since our last visit, had lowered the level of the ground water. 

Berkey and Morris decided that this was an excellent opportunity for a 
detailed study of a desert lake which would apply equally well, in generalities, 
to all those in this region. For this a map was essential, and Major Roberts 
with his assistants. Lieutenants Butler and Robinson, immediately set to 
work. The map was based on contour intervals of one and ten feet, the scale 
being one to twenty thousand. They found that the lake has an altitude of 
thirty-eight hundred and twenty-eight feet above sea-level. Seven beaches 
were apparent, the highest being thirty- three feet above the 1925 shoreline. 

The lake itself offered additional evidence of recent climatic fluctuations. 
In 1922, when we saw it first, Tsagan Nor covered an area of nearly six square 
miles, but probably nowhere was the water more than ten feet deep even then. 
The geologists believed that the major outlines of the depression could have 
been formed most consistently by stream erosion in a former more hiunid 
period, but the minor depression occupied by the lake must have been made 
by wind action at a time when there was no water in the basin. Therefore, 
there certainly was an interval considerably more arid than at present. The 
seven ancient beaches, high above the water level of 1922, could have been 
made only by waves, and indicate that the water stood at those respective 
levels long enough to form the strand lines. Hence, it is obvious that there 
have been periods of greater humidity than in 1922 and 1925. 

The geologists, Berkey and Morris, 1927, have remarked that: "If the 
lake were filled to the height indicated by the upper beaches, it would overflow 
at the east end and establish a stream which would follow a generally easterly 
course, marked out by a broad valley in that direction which doubtless was at 
one time the lake's outlet. 

"The reason for emphasizing this point, however, is its bearing on changed 
climatic conditions. At the present time, and apparently for a long while in 
the past, there has not been enough water ftimished to the Tsagan Nor basin 
to maintain a level of the lake high enough to overflow. In order to make the 
beach terraces, there must have been a time when water supply was more plenti- 
ful than now. This could happen only with a change toward greater htmiidity. 
Each of the well-marked beach levels must have been maintained for a long 
time. The changes that have taken place, therefore, have not been gradual or 
continuous, but they have come by steps or in cycles."^ 

1 Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, p. 392. 



June 10, the day we arrived at Tsagan Nor, was so hot that we sought the 
shade of tents and cars at every opportunity. The following morning heavy 
rain began at six o'clock, and a bitterly cold, east wind sent us into our fur 
coats. Such sudden changes of temperature come throughout the entire sum- 
mer, and we have learned that it never is safe to leave camp for a day's 
trip without warm clothing, even during the hottest weather of Jioly and 


When the various branches of work were well started at Tsagan Nor, 
Shackelford and I went out for motion pictures with the camera strapped in 
the tonneau of the car on its tripod. The hard, gravel plain north of the lake 
swarms with wild asses and antelopes. Seven miles from camp we stopped on 
the edge of a wide shallow depression. Even with the naked eye we could see 
hundreds of yellowish forms swimming in the desert mirage: wild asses, with- 
out a doubt, but never before had I seen a herd so vast. They were massed 
in three dense groups on the valley floor, and for miles the horizon was dotted 
with stragglers. By counting a block of two hvmdred we could estimate 
fairly accurately that there were at least a thousand animals in the herd. Sub- 
sequently, we learned that there were many more than that, for several hun- 
dred were below our sight in the bottom of a shallow ravine. 

It was obvious that we must circle far to the east and drive the herd west- 
ward onto the gravel plain where for fifteen miles we had splendid going for the 
car. When we finally headed toward the asses, a group of forty surprised us 
before we reached the main herd. The animals began nmning rather slowly, 
stopping often to gaze curiously at the car. We did not press them, but main- 
tained a steady pace of twenty miles an hour. The asses kept their distance 
ahead of us easily enough, and others began to come in from every side. The 
car drew them like a magnet. The yellow stream of asses converged on a 
course which brought them in front of us in a thundering mass. Hundreds 
were pounding along on both sides and we were enveloped in such a cloud of 
dust that I could barely see to drive. Photographs were impossible until we 
escaped from the dense herd of fleeing animals. I dropped back and swung to 
the outside of the largest group. There were perhaps three hundred asses so 
closely packed that they were running head to tail. As we shot forward beside 
the leaders, the herd swimg like a troop of cavalry and tried to cross our 

Shackelford, braced on the rear of the car, ground off film, swinging his 
camera from side to side as the mass divided. The fleetest individuals kept 
together, but the slower dropped behind, separated into groups of ten to 


thirty, and took their own courses. But soon the small herds began to 
converge again and so the performance was repeated. By the time we had 
reached the western end of the plain which drops off steeply into the wide 
valley of the Tatsin Gol, asses were pouring over the rim like a cataract of 
yellow water. 

When we turned back we encountered stragglers running about with their 
noses held high in the air, trying to find their companions. Yet they could not 
resist the fascination of the car; before we had gone a mile, more than a hun- 
dred were pounding along in front. Fifteen or twenty gazelles joined the 
parade, running with stiff black tails erect, sometimes springing into the air 
as if they were on pneumatic tires. 

I let the herd cross our bow and swung away to the south. With the glasses 
we covild see nine grazing quietly two miles away. We decided to try a new 
plan. Running a few yards, we stopped the car. The asses looked up, trotted 
toward us and stood with ears erect. Again they advanced and again. Before 
long they were less than two himdred yards away and Shackelford could work 
with the telephoto lens. This seems to be the only way to get pictures of the 
animals when they are not in rapid motion. It is impossible to stalk them on 
the plain for there is no cover. They do not often come to water-holes, as they 
seldom drink. One cannot successfully lie in wait for them anywhere, since 
they have the whole vast plain upon which to wander. 


Shortly after our arrival at Tsagan Nor, Granger and Olsen transferred 
their camp about fifteen miles due north, at Loh, where we had found the 
Baluchitherium skidl in 1922. We were exceedingly desirous of finding addi- 
tional parts of this huge aberrant rhinoceros. In a great deposit of red sedi- 
ments west of Loh, which we called the "Grand Canyon," Granger found the 
femur of a Baluchitheriwn, as well as hundreds of tmidentifiable bone frag- 
ments, evidently a skeleton which had been destroyed by weathering. Great 
quantities of rodent jaws and teeth, carnivores and insectivores, and a few 
pieces of fossil wood, were also yielded by this same formation. 

The credit for the most interesting discovery at Loh belongs to one of our 
Chinese collectors, Liu Hsi-ku. His sharp eyes caught the glint of a white 
bone in the red sediment on a steep hillside. He dug a little and then reported 
to Granger who completed the excavation. He was amazed to find the foot 
and lower leg of a Baluchitherium, standing upright, just as if the animal had 
carelessly left it behind when he took another stride. Fossils are so seldom 
found in this position that Granger sat down to think out the why and where- 
fore. There was only one possible solution. Quicksand! It was the right 
hind limb that Liu had found; therefore, the right front leg must be farther 


down the slope. He took the direction of the foot, measured off about nine 
feet and began to dig. Sure enough, there it was, a huge bone, like the trunk 
of a fossil tree, also standing erect. It was not difficult to find the two limbs of 
the other side, for what had happened was obvious. When all four legs were 
excavated, each one in its separate pit, the effect was extraordinary. I went up 
with Granger and sat down upon a hilltop to drift in fancy back to those far 
days when the tragedy had been enacted. To one who could read the language 
the story was plainly told by the great stumps. Probably the beast had come 
to drink from a pool of water covering the treacherous quicksand. Suddenly 
it began to sink. The position of the leg bones showed that it had settled 
slightly back upon its haunches, struggling desperately to free itself from the 
gripping sands. It must have sunk rapidly, struggling to the end, dying only 
when the choking sediment filled its throat and nose. If it had been partly 
buried and died of starvation, the body would have fallen on its side. If we 
could have found the entire skeleton standing erect, there in its tomb, it would 
have been a specimen for all the world to marvel at. 

I said to Granger: "Walter, what do you mean by finding only the legs? 
Why don't you produce the rest?" "Don't blame me," he answered, "it is all 
your fault. If you had brought us here thirty-five thousand years earlier, 
before that hill weathered away, I would have had the whole skeleton for 
you !" True enough, we had missed our opportunity by just about that margin. 
As the entombing sediment was eroded away, the bones were worn off bit by 
bit and now lay scattered on the valley floor in a thousand useless fragments. 
There must have been great numbers of baluchitheres in Mongolia during 
Oligocene times, for we were finding bones and fragments wherever there were 
fossiliferous strata of that age. 


While Granger and Olsen were at Loh, the geologists and Doctor Chaney 
went northward twenty miles, to Mount Uskuk, where we had discovered a 
dinosaur skeleton and a deposit of paper -shales in 1922, in the Ondai Sair 
formation. More fish and insect fossils were obtained, and Chaney got some 
rather poorly preserved plants. They were mostly conifers related to, but 
unlike, those now living. Doctor Chaney remarks that the shales are thin- 
bedded and appear to have been deposited in lakes, on the shores of which 
grew the trees. Their leaves and stems were blown and washed into the lake 
and there mingled with the Equisetutn and other rushes living in the water. 
The climate may be supposed to have been rather cool, with moderate or lower 
rainfall. If this had been a region of heavy rainfall, the variety and abundance 
of the fossil plant remains would have been much greater, as is the case with the 
Cretaceous floras in America and elsewhere. 



After the palaeontologists and geologists had finished investigating the 
region north of Tsagan Nor, they transferred their operations across the lake 
to the slopes of Baga Bogdo. A little later, aU the other men except Roberts, 
Young and I followed them. As far as we were aware, the snow-covered peak 
of Baga Bogdo never had been ascended by a white man. The Mongols them- 
selves have a fear of the peak; they believe it to be inhabited by fierce beasts, 
and we were told that no one who attempted to ascend could retiim alive. 
Bleeding at the nose would begin and continue vintil the person died. For this 
reason it seems improbable that the natives have gone to the summit, although, 
as mountain-climbing goes, it is not difficult. 

Doctor Loucks and Lieutenant Butler ascended Baga Bogdo on Jime 24. 
They left Tsagan Nor in the morning of the twenty-third and that night 
slept at the top of the alluvial fan, in the first gorge to the west of Tiger 
Canyon. The following morning was rainy and the entire motmtain was 
enshrouded in a thick blanket of fog, but they left at six o'clock and reached 
the summit at one o'clock in the afternoon. The ascent out of the head 
of the canyon to a ridge which runs to the base of the peak was the most 
difficult. From the western side the moimtain could be climbed much more 

They remained at the svimmit only fifteen minutes, for it was bitterly 
cold and the clouds enveloped them so closely that it was impossible to see 
more than a few yards. After planting the American flag and that of the 
New York Explorers' Club on the highest point, they built a cairn of rocks 
and left a letterhead of the Central Asiatic Expeditions in a bottle. They found 
no evidence of other men having been there ; had any of the Mongols ascended 
the peak they would almost certainly have built an obo. 

Baga Bogdo would doubtless prove to be an excellent shooting ground for 
sheep and ibex, as the animals are not hunted there by the Mongols. The 
natives assured us that tigers and leopards infested the moiuitain, but there are 
certainly none of the former within many himdreds of miles, and I never had 
authentic information of the latter in this region. 


Some distance from the lowest slopes of the moimtain the palaeontologists 
investigated fiuther the Pliocene deposit of gray clays which we had discovered 
in 1922 and named the Himg Kureh formation. I have already described its 
fauna in Chapter XIII, and their search did not greatly increase the list of 
species. Most interesting was the pelvis of a mastodon which measured sixty- 
four inches in diameter; also a great quantity of Struthiolithus shell fragments 
in a single deposit. Granger believes that this represented a nest of the gigantic 


ostrich. In the same deposit, Chaney found niimerous impressions of Equise- 
tiim and other swamp plants. Prior to the folding and uplift of the Altai, these 
plants grew on the border of a small lake. The beaver {Castor) substantiates 
such a view. The large deer certainly must have lived in a forest area, but 
the horse and ostrich were both adapted to the open plains. It is quite proba- 
ble that during the Pliocene this was an open savanna country, with lakes and 
patches of forest interrupting the meadowland. 


The ostrich must have been a very common and striking inhabitant, in 
late Pliocene and early Pleistocene times, not only of Mongolia but of all 
north and central China. Pere Emil Licent and Pere Teilhard de Chardin 
fotmd, at their palaeolithic site on the southern edge of the Ordos desert, that 
ostrich eggs had been an important article of food for the primitive humans. 
In the loess of China, Stnithiolithiis eggs are so frequently discovered that it is 
possible to buy them in many curio shops of Peking. The only known skeletal 
parts of the bird are a few toe bones which we foimd at Hung Kureh. Doctor 
Chaney had collected a hundred species of plants new to his collection at Baga 
Bogdo and brought back as a camp pet, the baby black vulture which I have 
described in Chapter XI. 


The geologists had an opportunity to study carefully the Baga Bogdo 
range which is about fifty miles long and not more than twelve miles wide. 
Baga Bogdo means "Lesser Buddha" as distinguished from Ikhe Bogdo, or 
"Greater Buddha," immediately to the west. The highest of the twin peaks 
of Baga Bogdo rises nearly eleven thousand feet above sea-level and about 
seven thousand feet above Tsagan Nor. It is one of a series of more or less 
isolated and recently uplifted mountain blocks which form the eastern extension 
of the great Altai system. Baga Bogdo is of the fault-block type of mountain 
which owes its relief to direct uplift or tilting. The geologists, Berkey and 
Morris, 1927, remark that "although the internal structure of the fault-blocks 
is as complex as the oldrock floor and includes folded rocks, the mountains 
themselves are not folded moiintains. . . . The folding and the granite 
intrusions were accomplished long before the uplift of the present fault-block 
range began." 

The northern face of the moiintain is cut into a series of deep canyons 
fronted by enormous alluvial fans, some being ten miles or more in length and 
two or three thousand feet high. The fans are the result of stream action dur- 
ing periods of greater or less moistiu-e and furnish much interesting evidence of 
climatic fluctuations. 



In trenches which have been cut by river action through the alluvial fans 
during comparatively recent times, cottonwood trees have found sufficient 
moistiire to maintain themselves. There is a considerable grove on the alluvial 
fan at the mouth of Tiger Canyon which represents a new species and has been 
described by Dr. Alfred Rehder, 1927, under the name Populus pilosa. The 
occurrence of trees in the desert parts of Mongolia is so infrequent and signifi- 
cant of climatic conditions that I am quoting in full Dr. Ralph W. Chaney's 
supplementary notes to Dr. Rehder' s paper: 

"The scarcity of trees in the Gobi desert region is striking evidence of the 
low rainfall over this great plateau. Elms, Ulmus pumila, are nimierous on 
the grasslands bordering the Gobi to the south but have been noted in only a 
few cases extending northward for a short distance into the desert proper. A 
single willow tree, Salix viminalis var. splendens, was seen in one of the valleys 
at Ondai Sair. But apart from these, no trees have been noted on the Mon- 
golian Plateau outside of the canyons of the Altai Mountains, a range which 
extends in a southeasterly direction across the western side of the Gobi desert. 

"The comparative abvindance of trees in the canyons of the Altai Moiui- 
tains is the result of the great precipitation there, and the higher degree of pro- 
tection from evaporation by the winds which are so characteristic of the Gobi 
proper. Not only are trees more abvmdant, but plants of all sorts are more 
nimierous and, as observed during the summer of 1925, continue in a green 
state long after the vegetation of the adjacent lower country has become dry. 
We experienced rain on three of the six days spent on Baga Bogdo during the 
latter part of Jxine, and there was a considerable fall of snow on the peak on 
June 20th ; a month later at Artsa Bogdo there were showers on foixr of the 
five days we spent in the mountains. Several of the higher peaks, such as Baga 
Bogdo and Ikhe Bogdo, have snow on their tops and protected slopes during 
most or all of the year. This was the case in 1925, and there was ample evidence 
to indicate that snow had persisted for at least two years in some of the larger 
canyons. As a result there are permanent streams in these canyons, along 
which conditions for plant growth are in striking contrast to those of the arid 
open slopes beyond the canyon mouths. None of the streams were observed to 
flow beyond the mouths of their canyons before they disappeared by evapora- 
tion and by sinking into the coarse gravel and sand of the fans. It was in these 
canyons and in the upper portions of the fans below their mouths that Populus 
pilosa was collected and observed on Baga Bogdo and Ikhe Bogdo. None were 
seen on Artsa Bogdo and Gurban Saikhan, the easternmost mountains of the 
Altai which extend farther out into the Gobi and may be supposed to present 
less favorable conditions for tree growth. The occurrence in one of the larger 
canyons on the north side of Baga Bogdo, called Tiger Canyon by members of 


the Expedition, will here be described as typical of the several similar occur- 
rences on this mountain and on Ikhe Bogdo to the west. 

"Tiger Canyon in its lower portion is cut into a coarse alluvial deposit, the 
walls rising steeply some 400 feet to an upper terrace ; its width is 2000 feet at 
the top and a few hiuidred feet at the bottom. A quarter of a mile above the 
mouth the canyon is cut into granite and is greatly narrowed with much 
higher walls. A mile and a half above the mouth metamorphic rocks form a 
still narrower canyon. The floor is littered with coarse gravel and with bould- 
ers up to 20 feet in diameter, and there are niunerous terraces which give it an 
irregular surface. The stream was at most only a few feet in width and less 
than a foot deep as observed in June, 1925. It has a high gradient and the 
water is clear and cold. Extending for several miles down the side of the 
mountain below the mouth of the canyon is a broad alluvial fan, cut by numer- 
ous dry channels, and littered with gravel and coarse boulders. The stream 
disappears into the gravel more than a quarter of a mile above the canyon 
mouth, but its presence in the gravels imder the surface of the fan may be 
inferred from the distribution of trees for more than a mile down the steep 
slope of the fan below the mouth of the canyon. These trees, all of which are 
cottonwoods, Popiilus pilosa, are from 15 to 25 feet in height and from 8 to 18 
inches in diameter. At the lower end of their distribution (elevation 5,200 feet) 
many are dead, indicating that the supply of water there is inadequate. 
Except for the trees the fan is essentially bare of vegetation, but along its 
borders a species of Artemisia, as yet undetermined, is abundant, together with 
several species of grass and legiunes, and low bushes of Primus mongolica. 

"In the canyon the trees are more numerous and larger, reaching a maxi- 
mum height of about 40 feet and a diameter up to 30 inches. Here the added 
protection of the canyon walls permits a more symmetrical growth of the trees. 
They are found along the stream for a distance of at least two miles up the 
canyon to an elevation of about 7,500 feet where it becomes too rocky and nar- 
row for them to gain a footing. Since most of the specific determinations of 
the flora have not yet been made, a complete and exact list of the associated 
plants cannot here be given, but it includes Salix phylicijolia which reaches the 
dimensions of a small tree, Salix glaiica, Cotoneaster melanocarpa, Lonicera 
microphylla, and Spircea chaniczdry folia among the woody plants, the fern 
Cystopteris fragilis, and numerous herbs of which the Leguminosas and Ranun- 
culaceae are well represented. No weedlings of Popidus pilosa were observed in 
Tiger Canyon or on its fan, but there were a few in the next large canyon to the 
west where there is also a permanent stream. While trees here are numerous, 
few of them reach the size of those in Tiger Canyon, the average diameter 
being little more than eight inches. No exact count of the trees was made, 
but it may be conservatively estimated at several htmdred in each of these 








The base has already been partly buried by drifting sand. 


two canyons. In the only large canyon observed at Ikhe Bogdo, the next 
large range west of Baga Bogdo, the trees were not as numerous and did not 
extend as far down the fan. 

"The use of the wood of Populus pilosa by the Mongols is abundantly 
indicated by sawed stumps in Tiger Canyon. Portions of the logs are hollowed 
out and made into tea mortars and water containers. The Mongol name for 
this tree, 'Tore,' meaning hollow, is indicative of this utilization of it by a 
people whose nearest approach to contact with forests is in the scattered 
groves of the Altai canyons. 

"In addition to Populus pilosa and two species of Salix, which were the 
only trees seen by the writer in the Gobi region proper, an arborescent species 
of Betula was seen by Charles P. Berkey in a large canyon on Ikhe Bogdo. It is 
significant to note that all three of these genera, as well as Ulmus of the grass- 
lands and Gobi border of the south, have windbome seeds, a fact which is 
probably responsible in large part for their distribution in the more suitable 
areas of the arid plateau of Mongolia. The source of the seeds which first 
established the trees of the Altai canyons may be supposed to be the higher 
and moister continuation of the range to the west. An alternative explanation 
may be that these patches of trees represent relict areas of a forest which was 
once more widespread and probably continuous with that farther west along 
the Altai Moimtains. The finding by Nels C. Nelson of birch-bark utensils in 
a prehistoric burial north of Ikhe Bogdo is evidence, in any case, of the presence 
of Betula in the region for several himdred years. 

"No data are available as to the age of the trees of Popuhis pilosa, but in 
view of their probable slow growth it may be supposed to reach several scores 
of years in the case of the larger individuals. Bearing on this problem is the 
situation of several of the trees near the head of the fan at Tiger Canyon. 
The basal 8 or 10 feet of their trunks has been buried by gravel deposits, which 
may be interpreted as indicating a fluctuation of rainfall and therefore of 
deposition during the period in which they have been living."' 


At the base of Baga Bogdo there were a number of circular mounds 
enclosed by stones which Nelson believed to represent graves. He opened 
half a dozen without finding anything of importance. At last he obtained a 
fairly well-preserved human skeleton but no implements. These graves doubt- 
less represent a pre-Mongol people of the Bronze and Iron Ages. 

' Rehder, Alfred. 1927. "A New Poplar {Populus pilosa) from the Eastern Altai Mountains." With 
Supplemental Notes on the Distribution and Habitat. By R. W. Chaney. Amer. Mus. Novitates, No. 292, 
Nov. 30, pp. 3-8; Preliminary Reports, Central Asiatic Expeditions, Amer. Mus. Nat. Hist., 11, No. 79, pp. 1-8, 



From Tsagan Nor we could see the mountain Ikhe Bogdo bulking hugely 
against the western sky. We knew that there was a large lake, called Orok Nor, 
at its base, but none of us had seen it. As we expected to continue westward as 
soon as our study of the Tsagan Nor region was ended, Loucks, Young, Shackel- 
ford and I went out to explore the district. 

We found that the hard gravel plain extends for thirty miles as an 
unbroken floor except where it is cut by the Tatsin Gol which flows into Tsagan 
Nor from the north. The bordering plain extends southward to the eastern 
end of Orok Nor and leads to an old beach, raised high above the water level, 
forming a wide semicircle six or eight miles long and at least two hundred feet 
wide. It formed a splendid gravel speedway on which the car could run at 
fifty miles an hour. The beach ended abruptly in a vast sand-dune area which 
encircled the lake except on the southern side, where the water lay close to the 
base of the mountain. 

We got the car through the dunes to the water at the eastern end after a 
hard struggle and much pushing, but reached only a series of reedy lagoons 
and not the main lake. The inlets swarmed with birds; eared grebes, coots, 
ducks of half a dozen species, geese, herons, storks, gvUls, terns, and many 
shore-birds were breeding there. I had never seen such a concentration of avian 
life anywhere else in Mongolia. 

It was evident that the lake was many miles in length and very wide. 
Some hundreds of years earlier it must have been of vast extent — a veritable 
inland sea. The area now covered with yellow sand-dunes must have been 
under water in quite recent times. At least two large rivers flow into it from 
the north, although they contained little water at the time of our visit. 

After spending the entire day in a fruitless endeavor to find a way for the 
cars to the main shore, we camped at sunset where one of the rivers breaks 
through the hills on its way to the lake, nine miles away. Our sleeping-bags 



were spread on a grass-covered shelf beside the river, just under the brow of a 
rocky cliff. At dusk, bats of two species darted in and out among the rocks, 
and we did oiir best to collect a few specimens, but as they appeared for only 
the fraction of a second against the sky it was like shooting at a phantom. 


After another unsuccessful attempt to reach the lake next morning, we 
turned north and climbed in the car to the summit of an exceedingly rough 
range of mountains, from which we could look back and see the western end 
of Ikhe Bogdo. We thought that we could pass quickly across the mountain 
summits to the northern face. To our great surprise the range was eight 
miles across. Each summit was connected by ridges just wide enough for the 
car to pass. Expecting to reach the other side at any moment, we kept on 
tmtil we were deep in the maze of peaks. It was a strange place for a motor 
car! We were about two thousand feet above the level of the plain, surroimded 
by rugged crags. Because of Young's skilful driving we eventually reached the 
northern face of the mountain, emerging by way of a narrow rocky canyon. 

In the distance we could see gray and red bluffs and isolated buttes in 
half a dozen places, but the enormous basin in which they lay was deep in 
loose sand and spotted with "nigger-heads." We reached several of the ex- 
posures after hard work, and although few bones were to be found, they evi- 
dently contained some fossils — certainly enough to warrant more careful 
investigation in the future. 

At last we came to what we eventually learned was the old Uliassutai 
caravan trail which passes north of Tsagan Nor, and followed it eastward. 
We were glad to be on passably decent terrain again. Thirty-eight miles 
before we reached Tsagan Nor the trail led us into a low marshy bottom 
from which we could see two lakes some distance to the north. By retiring to 
the high groiuid and making a wide circle, we reached the farther and smaller 
lake, which was evidently rapidly drying up. The mud of its shores gave 
forth such a strong odor of carbon bisulphide that it was nauseating. It had 
even driven away from the green grass of its shores the several herds of sheep 
and goats which were grazing in the valley. 

While we were on our way to this lake, a violent storm arose. The wind 
blew from the north with such velocity that it was almost impossible to force 
the car against it. At last it became so strong that the car could do no better 
than three or iovur miles an hour even with the engine running at full speed. 
We turned back just as a terrific deluge of rain began; in five minutes we 
were shivering even though our rubber coats kept us fairly dry. 

The larger lake, which we learned was named Kholobolchi Nor, proved to 
be a charming place. To the west the gravel plain dropped abruptly down 


to the lake hollow where a beautiful green lawn spread right to the water's 
edge. It would make an excellent spot for a base camp from which to investi- 
gate the surrounding region. Returning to the trail we continued eastward 
toward Tsagan Nor. After going only six miles the trail descended to a wide 
basin filled with huge "nigger-heads." At the western edge a clear spring ran 
out of a low rounded bluff of gray clays. After a few moments' search we dis- 
covered a few bits of fossil bone, but did not remain to make a thorough search. 


The Expedition moved to Kholobolchi Nor June 28. Tents were pitched 
on a strip of turf as green as emerald. Mine was not more than two feet from 
the water's edge, at a point where it cut under a low bank. Immediately in 
front of the door we dug a well, or rather two wells. Shackelford and I were 
sponsors for one ; Olsen and Loucks for the other. Each of us sank a gasoline 
box in the pit and surrounded it with white stones. The water was clear and 
sweet, for the entire grassy terrace was underlaid with springs, the water 
being rather too close to the surface. 

Nineteen white swans were floating quietly near shore when we drove in 
and moved only a few hundred yards down the lake when our tents were 
pitched. A great black and white stork, Ciconia nigra, with red legs and beak, 
stepped majestically along the water's edge; gulls and black-capped terns 
fished energetically, and dozens of sheldrakes trailed their tiny broods behind 
them like waving streamers. 

On the desert we had seen one or two brown vipers, Agkistrodon, but 
near the lake I caught a new snake, Coluber spinalis, which Cliiford Pope, 
1929, records also from Shantung, Chihli and Shansi provinces, China. 

After tiffin the entire camp went in for a swim. The bottom is fine sand 
in most places, but one must wade out for two hundred yards before the 
water reaches one's shoulders. To the west and south, sand-dunes, which 
are an invariable accompaniment of all Gobi Desert lakes, lie in yellow waves. 
Three or four Mongol yurts were pitched just at their edge. That night we 
were treated to a glorious sunset ; as we watched the gorgeous display, mean- 
while listening to the victrola, all of us agreed that an explorer's life is not 
so bad. 


On the way to Kholobolchi Nor, Berkey and Morris discovered a few 
fossils in the gray clays six miles from the lake which I had "marked down" 
on our first visit. Mastodon and horse, Eqiius, could be readily identified, 
indicating that it was a Pleistocene deposit. The next day the palaeontolo- 
gists prospected the exposure more thoroughly and discovered some good 


specimens of teeth and jaws, confirming the view that the deposit was of 
Pleistocene age. We were particularly glad to find such late sediments, for 
they had been absent in the region we had explored previously. The Pleisto- 
cene appears to have been a period of erosion rather than of deposition in the 
central and western Gobi. 

Not far from the Pleistocene bluff, near Kholobolchi Nor, Doctor Berkey 
discovered a thin patch of Eocene sediments in which Buckshot found the 
skull of an Amblypod, probably related to Eudinoceras. Olsen discovered a 
second skidl which appears to represent a different species. Except for the 
two teeth obtained in 1923 by Professor Osbom and myself at Irdin Manha, 
this American group was tinknown in Asia. Although the two teeth were 
immistakable and definitely placed the great group of amblypods as former 
inhabitants of Asia, these skulls will tell a more complete story of their rela- 
tionships to the American forms. 

Granger went many miles in the car, investigating every likely-looking 
spot in the huge basin. As a whole, the Pleistocene deposits and exposures 
of Oligocene age north of the lake proved rather disappointing, for the fossil 
content was neither rich nor varied. Still, what we did obtain was of the 
highest importance. 


Nelson found more of the Dune Dweller culture in the "tamarisks" of the 
basin just in front of the Pleistocene exposure. Also we were thrown into a 
fever of excitement by Nelson's discovery of a human skeleton in the bluff 
itself one evening just at dusk. Ftuther excavation proved it to be a burial, 
much to our disappointment. Some of the bones were wrapped in birch 
bark and pieces of wood had been used to roof the grave. Without question 
it was pre-Mongol and more than a thousand years old, but we had hoped 
for a skeleton of Pleistocene man. Subsequently, Nelson foimd a consider- 
able exposure of the Shabarakh formation near Orok Nor, lower than the 
ancient beach levels. Thus the water had covered that formation for a long 
time after the Dime Dwellers had vanished. This bears out the geologists' 
conclusions that the ctdtiu"e is many thousands of years old. 

On the gravel plain east of camp. Nelson found several artifacts which 
could safely be identified as of late palaeolithic technique. They included 
choppers or large scrapers and oblong flakes of Mousterian type, as well as 
double-end scrapers of Aurignacian workmanship. Although none were ac- 
tually found embedded in the Pleistocene plain, nevertheless there was strong 
reason to believe that the artifacts belonged there and had weathered out of 
the strata. The area has been undergoing erosion probably ever since mid- 
Pleistocene time. 


On the same surface and in a number of other localities, especially in the 
region between Tsagan Nor and Orok Nor, there are thousands of fractured 
boulders and pebbles more or less weathered but showing successive stages of 
flaking. As Mr. Nelson says: "A large number of these splintered pieces 
assume artificial form and character, even to the extent of now and then 
showing the bulb of percussion. In addition, nearly all of them exhibit more 
or less retouched margins, a succession of chips having been removed much 
after the manner of true Mousterian technique. Yet the obviously varying 
ages of the chip beds leave no doubt that the specimens are the veritable 
product of natural forces still at work."^ 

I must state that Mr. Nelson takes very little stock in eoUths as a whole 
and those of Mongolia in particular. After walking for some miles behind 
our caravan, when it was passing over a plain strewn with jasper rocks and 
large pebbles, he came to the conclusion that a camel with its feet could 
make just as good eoliths as could primitive man with his hands. Rocks 
were broken and flaked by the weight of these heavy animals in a way which 
imitated most remarkably the eolithic technique. I believe I am correct in 
stating that, after his experiences in Mongolia, Mr. Nelson would regard 
suspiciously the finest eolith that ever has been discovered. 

Nevertheless, much remains to be elucidated regarding the true and false 
artifacts of the Orok Nor region. It certainly is the most important locality, 
from the standpoint of a very primitive culture, that we discovered in Mon- 
golia. There is great probability that himians using the Mousterian tech- 
nique lived during the Pleistocene in this locality. Prolonged search might 
produce much more evidence of their habitation. In a region such as this, 
where there is no hope of finding caves or rock shelters, archaeological work is 
exceedingly difficult. Our greatest hope of discovering human fossils lay in 
river drift or bog deposits. The fact that the Jesuit Fathers, Teilhard and 
Licent, found a great deposit of Mousterian implements in the Ordos desert, 
just south of this region, is most important. Their discovery could be defi- 
nitely dated by accompanying animal bones, and Nelson believes that his 
Mousterian-type implements represent the same culture. 

The deposit found by the Jesuits was on the shore of an ancient lake, 
long since "drowned" by drifting sand. It is probable that in Asia Neander- 
thal man, or his contemporary, was a lake-shore dweller. Doubtless he sought 
the protection of a bank, or he may have built himself a shelter of branches 
roofed with skins. There must have been just as few caves and rock shelters 
in his time as there are at present. 

' Berkey, Charles P., and Nelson, N. C. 1926. "Geology and Prehistoric Archasology of the Gobi 
Desert," Amer. Mus. Novilates, No. 222, June 28, p. 11; Preliminary Reports, Central Asiatic Expeditions, 
Amer. Mus. Nat. Hist., 11, No. 67, pp. 1-16, 1930. 



Two days after our arrival at Kholobolchi Nor, Chaney, Shackelford and 
Loucks went to one of the lagoons of Orok Nor to photograph and botanize. 
They found hundreds of water-fowl breeding on the islands of tide grass, but 
they were about two weeks late for nests. Much to their surprise, a great 
flock of white spoonbills frequented the lake. This was the first time we had 
encountered them anywhere in Mongolia. 

Chaney obtained a splendid collection of plants. He found that the 
flora is much like that of our American lakes, including water buttercups, 
bladderwort, pond weed, green algas, cattails, tule and duck weed. Strangely 
enough the arrow leaf, pickerel seed, bulrushes and several others of the 
American lake flora are absent. 


The geologists made an arduous but profitable trip to Ikhe Bogdo, spend- 
ing several days on the mountain. Their principal object was to investigate 
the glacial cirques which could be seen in the heads of all the major valleys 
and which record the easternmost glaciation in the Altai system. The geol- 
ogists have said that, except for the cirques found in several mountain ranges, 
the Ice Age in Mongolia is the most difficult period to read, because of the 
lack of recognizable deposits. Nevertheless, it is evident that the climate of 
the Pleistocene was more variable than any post- Cretaceous period; that it 
included epochs warmer than the present and epochs in which the higher 
mountains bore large alpine glaciers, while parts of the Siberian coastal slopes 
were covered by broad ice-caps. But the Gobi region itself was not glaciated, 
the ice being confined to the actual mountain valleys. The valleys suggest 
that the Pleistocene in Mongolia was largely a period of erosion, for it is 
evident that an enormous amount of material was removed from the country 
in geologically recent times. 


Berkey and Morris, 1927, also remark that the "peculiarly characteristic 
Pleistocene deposit called loess seems to be lacking in Mongolia. In our 
explorations we saw little or none, and we find no reports of it in the literature 
except from the extreme southern margin of the Gobi. The loess of China, 
south of the Gobi, may well have been blown from Mongolia and from the 
basin of Lop Nor, especially if the mountain barriers that now separate these 
lands from China were lower during part of Pleistocene time than they are 

' Berkey, Charles P., and Morris, Frederick K. 1927. "Geology of Mongolia," Natural History of 
Central Asia, Amer. Mus. Nat. Hist., II, p. 385. 


The problem of the origin, climatic conditions and age of the loess in 
China is one about which there has been much discussion. It is a fine-grained 
loam formation which is by no means restricted to China but is fotmd in 
various parts of the Northern Hemisphere. The term "loess" was applied 
to it in the German Rhineland. It is a soil of high fertility which covers a 
large part of north China in some places to a depth of two hundred feet. 

Barbour remarks that "the true Chinese loess is a yellow-grey, poorly- 
consolidated loam deposit of the fineness of silt, which shows a characteristic 
absence of horizontal layer structure, being essentially non-stratified and 
showing a tendency to split along roughly vertical joint-planes so as to form 
perpendicular cliffs and walls. "^ 

There are very strong reasons to believe that wind has been the principal 
agent in the formation of loess. It contains few fossils, but Dr. J. G. Anders- 
son has recorded finds of rhinoceros, sheep, hyena, bear, beaver, horse and 
deer. The eggs of the giant ostrich Strutkwlithus are frequently discovered in 
the loess of China. It is quite probable that these were covered with a wind- 
drift deposit and thus prevented from hatching, as undoubtedly happened in 
the case of the dinosaur eggs of Shabarakh Usu. It is a current remark in 
Peking that the disagreeable storms which aflect all north China in the spring 
bring "dust from the Gobi Desert." We have already seen that there is little, 
if any, loess in the Gobi; the sand and gravel are too heavy to be carried for 
any considerable distance. As a matter of fact, most of the storms of Peking 
sweep up local dust from the streets and the immediate surroundings of the 
city. The instances when yellow clouds floating hvmdreds of feet in the air 
cover vast areas of north China are comparatively rare. During those storms 
a fine yellow silt sifts through the cracks of doors and windows and into the 
tightest boxes. This is obviously loess which probably comes from Chihli, 
Shansi, Shensi, and other north China provinces. These storms are quite 
unlike those that we have in the Gobi. 


While work went on at Kholobolchi Nor, my thoughts were busy with 
the future. I was anxious to see what lay beyond the Altai Mountains. Day 
after day I had gazed at the massive ramparts barring us from the south. 
The natives related tales of wild camels and of the famed Prjevalski horse; 
they told us of barren gravel deserts, of sand and mountains, of death from 
thirst. But each tale only strengthened that restless urge which every ex- 
plorer knows — the desire to go and see. The mountains lay there like a sUent 
challenge. We knew that we could cross them on ponies, but could we cross 

• Barbour, George B. 1925. "The Loess of China," Chitia Journ. Sci. and Arts, III, No. 8, pp. 454- 
463, and No. 9, pp. 509-519- 


with a motor car? We should never know until we tried. Kozloff, the famous 
Russian explorer, told me that he had crossed the Altai near this spot, but he 
had a caravan of camels. We thought that we had located the pass he used, 
for we could see a sharp break in the peaks just west of Ikhe Bogdo. On 
July I, Roberts, Yotmg, Lovell and I left camp in an automobile with my 
faithful Mongol, Tserin. We carried an assortment of spare parts, food for 
a week and gas to run five hundred miles. Granger knew the general direc- 
tion we intended to take and that our objective was to get through the moun- 
tains some way ; if we did not return he could trail us in another car. 


After running a few miles westward we headed directly south toward the 
mountains. Roberts, by taking compass directions, was roughly mapping our 
route. From the summit of a low rise we saw a small lake about two miles to 
the west. Gulls and terns were flying over the mirror-like surface and islets of 
txile grass stretched a long green finger toward the center. From our eleva- 
tion Roberts sketched the shoreline while I studied it through my powerful 
binoculars. Slowly it began to dawn upon me that something was wrong 
about that lake. The beach grew indistinct and the tule grass islands danced 
about in a most peculiar way. Roberts and Tserin both noticed it, too. 

"Bob, I think we had better run over there before you go any further 
with that sketch," said I, and started the engine. In five minutes we were 
on the "shore" of the "lake" — only there wasn't any shore and there wasn't 
any lake. It was the most perfect mirage we had ever seen. Not even a sug- 
gestion of water or of the tule grass islands, and our "gulls" were sand-grouse. 
Yet from first sight, all of us woiild have staked our lives that it was real. 
It is an axiom of Arctic exploration that you never can be certain that land 
is land until you have put your foot upon it. Cloud banks lying over the ice 
make perfect mountains and coast lines. It is an axiom of desert exploration 
that a lake never is a lake until you have waded into its waters. 


The mirage served a useful purpose, for during our investigations we had 
crossed a well-marked trail which led toward the foothills between us and the 
mountains' base. It took us up a dry stream-bed, across a grassy ridge and 
into another wash. In some places the gorge was wide with bare rocky slopes ; 
in others the stream had cut a narrow canyon, and sheer walls towered above 
us five himdred feet or more; sometimes great rock slides threatened to bar 
our way, but always there was a gate through which the car could slip. 

We emerged into a beautiful valley on the north side of the majestic 
ramparts of Ikhe Bogdo, the Great Buddha, its snow-covered peak rising 


into the clouds. Our trail led up a vast alluvial fan ten miles long, toward a 
deep cleft in the mountain wall. I realized that it must be a river gorge and 
probably would be choked with boulders. The fan gave promise of what 
was to come. A chaotic mass of rocks paved the surface, and it seemed mad- 
ness to drive a car into the debris. Still, Young and Lowell did pilot it safely 
for ten miles actually into the canyon's mouth. There we stopped and con- 
tinued on foot. From the summit of a thousand-foot cone we could see how 
the gorge wound in and out among the peaks, passable for horses or camels 
without a doubt but hopeless for cars. We named it "Kozloff Pass," as it is 
almost certainly the one the great Russian explorer discovered. 


At the western end of the beautiful valley north of Ikhe Bogdo the hori- 
zon dropped to a level ridge where the mountain chain seemed to break. It 
appeared to be not more than ten or fifteen miles at most, but we ploughed 
forty miles through the sand before the crest was reached. Then we dis- 
covered that the range bent sharply to the south and that higher and rougher 
peaks lay beyond. There was not a sign of human life, but a dry lake-bed ran 
the entire length of the valley, which swarmed with antelope and wild asses. 
They were feeding on alfalfa, and we found this plant growing wild at half a 
dozen spots in other parts of the Gobi. I never have seen such a concen- 
tration of game in a small area. Antelopes were running beside the car and 
crossing our course every moment; tiny fawns hardly larger than rabbits 
jumped out from almost under the wheels, where they had been lying flat on 
the ground with necks outstretched. 

Herd after herd of wild asses pounded along beside us, unable to tear 
themselves away from the fascination of the car. IMost of the asses were 
mares and many of them were chaperoning fuzzy, long-legged colts. It was 
amusing to see the little fellows bend to the work of keeping up with their 
mothers. With ears laid back and slim legs flying they put every ounce of 
strength and determination into what probably was the first time in their 
short lives that they had run from danger. Once we saw four wild asses 
fighting. Kicking and biting viciously, they kept at it imtil the car ap- 
proached and they joined the zoological assemblage which we were driving up 
the dry lake -bed. 

In spite of the thousands of animals, there was something utterly deso- 
late about the vaUey. Perhaps it was the black mountain walls which shut 
us in and the fact that for more than a hundred miles we had not seen even 
the remains of an old camp-fire or the circular mark left by a Mongol tent. 
All of us were exhausted when we camped at dark in a sandy stream-bed. 
The speedometer of the car registered one hundred and fifty miles, and in 


that entire distance not a well or stream had we seen. There was a gallon 
of water left in one of the bags which would do for drinking and coffee, but 
we did not worry, for in the old lake-bed half a dozen patches of vivid green 
grass indicated that water could not be far below the siuface. During our 
night in "Deserted Valley" it rained heavily. We had no tents, but, pulling 
the flaps of the canvas sleeping-bag covers over our heads, we remained 
perfectly dry; moreover, there was the comforting assurance of sufficient 
water in the morning. 

The day began with hard work. When crossing a dry stream-bed the 
car suddenly stmk to the hubs in moist sand and there it remained for iour 
hours. Experience has taught us to take such things philosophically. With 
hardly a word everyone began to unload and to collect stones. To build a 
rock foundation under the wheels is the only possible way to get out of such 
a predicament. The quicksand appeared to be bottomless and the stone 
base was six feet deep before it would hold the jack and the weight of the 

Across the valley there was, in the ragged line of peaks, a dip which sug- 
gested a pass. None of us had much hope that it would be possible to get 
through, but it was the only chance. Crossing the low foothills successfvilly, 
we started up the slope only to emerge from behind a rocky comer on the 
very brink of a stupendous chasm. Red granite ridges capped with dull 
black lava cut into a thousand fantastic shapes showed against a lowering 
sky. In the utter stillness it lay like a red inferno. While Roberts took the 
compass points for his map, the rest of us explored the nearest canyon, which 
divided into a labyrinth of passages and roofless corridors. I suppose that 
some day when a railroad parallels "Deserted Valley," tourists will picnic in 
the gorge. Of course, they will name it "Dante's Hole"! 


A long detour took us aroimd the chasm and the break in the saw-tooth 
horizon proved to be a pass indeed. A hard floor of gravel led gradually 
toward the summit between slanting peaks. It was only seven thousand feet 
high, but it seemed as though we were mounting toward the roof of the world. 
As the car swept upward we sang and laughed, our spirits soaring with every 

From the crest, a vast panorama of low ridges spread out before us like 
the waves of a great sea in a heavy gale. We could look far into the mys- 
terious region south of the mountains which for us at the moment was the 
"Land of Heart's Desire." But we soon foimd that its interest lay chiefly in 
anticipation. It was beautiful but commonplace. Great plains sloped gently 
downward and the car flew like a bird over the gravel surface. Not a sign 


did we see of the reported fossil "badlands" or the terrible desert of thirst 
and death ; only line after line of pink-white ridges of quartz and marble. 


We crossed a well-marked trail running east and west. Although on 
none of the so-called maps was there an indication of a caravan route, we 
discovered later that it was what caravan men call the Great Mongolian 
Road to Chinese Turkestan. We swung east on the trail and found splendid 
going. The great fiat pads of a camel's feet are natviral road-makers, tramp- 
ing the sand until it is as hard as rock. In spite of the Mongol reports of the 
lack of water, the trail led us to a magnificent spring and just beyond it we 
saw the blue tent of a great caravan. They were Chinese from Shansi Prov- 
ince. As I know that dialect and we were all wanderers in the desert, they 
greeted us like old friends. In the big tent we drank tea and ate boiled millet. 
Twenty men with two hundred camels, they were on the way to Kobdo near 
the northwestern frontier of Mongolia. It was early May when they left 
China and they would not arrive in Kobdo imtil January, 1926. Nine months 
of the same life day in and day out, making and breaking camp, eating and 
sleeping. Nothing to interrupt the dreary monotony except the winter's fight 
against snow and cold and perhaps a bandit raid. 

Tea, cloth and tobacco were their goods to barter for camel's and sheep's 
wool, hides, furs and ponies. The same trade in the same way over the same 
trails has gone on for vmtold centuries and will continue until that not far 
distant day when a railroad connects China with central Asia. Then at one 
blow the romance and glamour of the desert will be destroyed. Tourists will 
sit in heated cars, eating the food of Europe, reading week-old newspapers, 
and comprehending not at all the glorious history, the tragedy and the ro- 
mance of the Gobi trails. 


The Chinese could give us no late news, but we learned much about the 
country, for they had made this journey four years earlier. The wild camels, 
they said, were far to the southwest, just west of the Etsin Gol in the Black 
Gobi; the trail we were on eventually broke through the Altai Moimtains and 
swimg north to Uliassutai and Kobdo; for several htmdred miles both east 
and west the country was a gravel plain. Three of the men said that they had 
seen wild camels in the Black Gobi but only at a distance. They were very shy 
and almost like mythical creatures. Some Mongols, they said, had caught them 
when yoimg but that even then they were difficult to tame. As far as I can 
discover, very little has been published about wild camels. Doubtless this is due 
to the fact that few explorers have visited the arid Black Gobi where they live. 



Continuing eastward on the trail, we camped on the banks of a small 
but swift stream called the Tsagan Gol (White River) which runs southward 
from the foothills between Ikhe and Baga Bogdo. In 1926, when I went to 
London to lecture before the Royal Geographic Society, I examined, with 
Sir Francis Younghusband, the route maps he made of his famous journey 
from Peking to India in 1887. We found that he, too, had camped at this 
very spot. Strangely enough, my thoughts were full of this gallant explorer 
on that evening. I knew that we must be near the route which he followed, 
and I told my companions much about this trip which he had made when 
only twenty- four years old. We had a delicious dinner of antelope steaks 
cut from a yoimg buck I had shot the day before ; then we lay in our sleeping- 
bags smoking and looking up at the stars until nearly midnight. 

A few Mongol yiarts were pitched near the river, and an old native told 
us that another large caravan trail paralleled the one we were on, to the south. 
He also said that he had seen both wild camels and saiga antelopes about two 
hundred miles to the southwest in sandy country. This, I think, was a mis- 
take, at least as far as the saiga are concerned. The trail of which the old 
Mongol spoke we foimd ten miles south of the one we were on. At that point 
it ran almost east and west. Doubtless it eventually swings southward to 
Hami, after it has avoided the sand-dvme area which it encircles. 


The country was unsatisfactory from our standpoint. It was an ex- 
tremely arid desert. Except in a low basin near a small lake, called Nagan 
Nor, and at Tsagan Gol, we saw no Mongols. There was not sufficient vege- 
tation on the hard gravel plain to support a herd of camels. Perhaps a dozen 
gazelles and two or three asses were the only wild animals that we saw. Sev- 
eral of the wells on the caravan trail were dry. It would be a difficult country 
to cross except during an unusually wet season, and the trail gave evidence 
that it had not been used for a long time; probably seldom since 1921 when 
Soviet-controlled Outer Mongolia placed prohibitive taxes on passing cara- 
vans and the trade with Chinese Turkestan was diverted to the Winding 
Road south of the frontier. There was not a sign of exposures which might 
contain fossils. The topography consisted of a series of short, low mountain 
ranges paralleling the Altai system. The inter-movintain basins, fifteen to 
thirty miles in width, were filled with materials eroded from the motintains 
themselves, which were composed of pink and white marble, red granite, 
quartzite and slate. No ravines, gullies, bluffs or sedimentary deposits were 
to be seen. It was quite evident that we should have to go far to the south 
to reach low basins where sediments might be exposed. In the entire trip we 


were always above five thousand feet altitude. As gasoline was running short 
we retraced our way over the pass by which we had come and reached camp 
without diihculty. 


In the afternoon, McKenzie Young developed a high fever, and by the 
time we reached Kholobolchi Nor he was feeling very ill. His was the first 
case of a peculiar malady which attacked all of us with the exception of Olsen 
and Granger. Doctor Loucks said that the symptoms somewhat resembled 
those of influenza. He believed that it was induced by camping on the grassy 
shelf beside the lake, which was very damp, due to the presence of under- 
ground water close to the surface. This was the only time during all our work 
in Mongolia that any member of the Expedition, either native or foreign, 
was seriously ill. Young, although one of the strongest men of the party, 
had the most severe attack of the malady and was on the sick-list for two 
weeks. The fever left him extremely weak, as was the case with all of us. 


The Fovirth of July was a delightful day: perfect weather, not a breath 
of wind, and the lake like glass. Chaney, Loucks, Shackelford, Berkey and 
Morris were away, but the rest of us celebrated by loafing in the sun, reading, 
writing and bathing. At night a dinner for which I had provided before 
leaving Peking gave us something to remember all our lives. 


During the night of our return to camp a strange thing happened. We 
were awakened by fish! A strong wind blew from the west until about two 
o'clock in the morning, pushing the water in the shallow lake over to our side. 
Suddenly the wind dropped and the water receded so quickly that thousands 
of small fish which had been feeding close up under the bank were left stranded 
on a strip of mud about three feet wide. Flapping wildly as they tried to 
work back into the water, they made a noise like scores of people softly clap- 
ping their hands. 

When I was awakened by the noise close to my head, I stepped out into 
the brilliant moonlight to learn what was going on. Thousands of glittering 
forms were flashing along the shore. I called Granger, Buckshot and others 
of the Chinese. With oiu- twenty-foot seine they brought in htmdreds of 
fish at every haul. We had some fried for breakfast, but they were too soft 
and "muddy" to be very palatable. The Chinese, however, liked them im- 
mensely and spent hours salting and drying them in the sun for future use. 
Most of these fish were about eight inches long, but Shackelford brought 


back the head of one of the same species from Orok Nor which indicated an 
individual at least two feet in length. All the fish that the men obtained in 
Orok Nor were like those of Kholobolchi Nor. But the fish in Tsagan Nor, 
only thirty-five miles to the east and in the same drainage depression, are of 
quite a different species. This is strange, for Orok and Tsagan Nor must 
have been connected not many thousands of years ago. 

Probably the fish were brought in by streams that flow, or have flowed, 
into the lakes. Still, how these desert bodies of water were stocked with fish 
is not entirely clear. In some cases it may have been by means of birds which 
carried fish or eggs from one lake to another. Just what happens to the fish 
when a lake dries up and remains dry for some time is also puzzling. Tsagan 
Nor, which was well-stocked with fish in 1922, had dried up entirely on July 
16 when we returned. Would the fish all die and the lake have to be re- 
stocked, or woiild some of them manage to exist by burrowing into the mud? 
If the latter is true, they would have to go down a considerable distance, 
because the surface of the mud already had dried almost as hard as flint in 
many places and was deeply cracked. I imagine that Tsagan Nor could 
receive a fresh supply of fish from its inlet, the Tatsin Gol, when that stream 
revived sufficiently to fill the lake again. 


One day Granger and Olsen brought in thirty-two bats, besides a great 
mass of tiny naked "batlets." They had caught them with their hands in 
the crevices of a cliff. Also they had two young homed owls and two of a 
smaller species. But owls do not make interesting pets, and they soon found 
their way into the specimen boxes. The scarcity of bats in the Gobi is inter- 
esting. We saw only a very few and those at not more than three or four 
localities. Doubtless this is due to the desert dryness which is favorable to 
but very few insects. 



On July 10, Granger, Berkey, Lovell and I started westward from Kholo- 
bolchi Nor on a reconnaissance trip. Our object was to get a general view of 
the region at least as far as the longitude of Uliassutai, and explore the basins 
which lie parallel with the Altai Mountains on the north side. Even though 
there was every reason to believe that the basin sediments continued for a 
long distance, before I took the entire fleet westward I wished to be sure that 
there was opportunity for effective work; particularly, because we did not 
have sufficient gasoline to go forward on a fruitless expedition. To move 
our seven cars one mile required almost a gallon of gasoline iinless the terrain 
was exceptionally fine. Leakage of gas, as invariably happens, had been 
heavy, and the hot days followed by cold nights gave the worst possible con- 
ditions for carrying our motor fuel. 

Granger already had visited exposures thirty miles west of camp and, 
to our amazement, they proved to be almost the last that we found. The 
first day we traveled one hundred miles without crossing a fossiliferous ex- 
posure. Seventy-five miles from camp, we saw, far to the north, a line of 
red badlands, evidently an extension of the Oligocene, Loh formation, but 
the glasses showed them to be so thin and they were so difficiilt to reach that 
we did not investigate further. 

Roughly, the structure over which we passed was an extension of the 
great Mongolian Bathylith with inclusions of limestone and gneiss at certain 
points. The valleys were floored with sediments, but there were no expostu-es. 
The country was extremely uninteresting and monotonous. The vegetation 
consisted only of Gobi sage and a few low, thorny plants; no wild asses and 
few gazelles of either species. Even the sand-grouse seemed to have deserted 
the region, leaving it stark and dead. 

THE batdarik; gol 

We camped the first night at the Baidarik Gol, a large river which flows 

into the lake, Bovm Tsagan Nor. The lake itself is fifteen or twenty miles 



south of the trail and we did not visit it, but the natives said it was a con- 
siderable body of water. On the way we passed one other lake, about the 
size of Tsagan Nor but dry at the time, as well as several ancient basins with 
beautifully marked beach levels. This whole region must have been dotted 
with lakes and very well-watered by streams during the moist climatic cycle 
which preceded the dry epoch in which we are now living. 

Baidarik Gol runs in a wide valley and, where the trail enters it, divides 
into three branches. It is deep in places, but we crossed it with the car along 
the line of ripples without difficulty. In the river valley, and also farther to 
the west, there were signs of irrigation, indicating that there had been an 
attempt at agriculture in the not very remote past. Probably this was the 
work of Chinese who were stationed at this post. On the west bank, on a 
bluff overlooking the valley, is an ancient watch-house made of hard clay 
bricks which would resist weathering for a long time. About three miles to 
the south we examined the ruins of what evidently had been barracks. Dur- 
ing the Manchu rule in China this must have been an important station for the 
administration of the post road. It is reported that gold has been foimd at the 
Baidarik Gol in considerable quantities, and I believe that the Russians have 
done some work along the stream, but there is no evidence of recent operations. 

After making camp we examined the high west bank of the river. At a 
distance it appeared to be a fine exposiu"e of sediments but inspection proved 
that it is formed of very old gneiss. The valley is the home of hundreds of 
wild ducks, geese and sand-grouse. The many fire holes evidence that it is a 
favorite halting place for caravans boimd to Uliassutai and Kobdo. Indeed, 
two large Chinese caravans were camped on the east bank when we arrived. 
They were loaded with tea, sugar, cloth and tobacco, and would bring back 
sheep and camel wool, hides and ponies. 


On the second morning we stopped for tiffin at a collection of yurts beside 
a large pond. The Mongols told us that the trail divided there, one branch 
leading north to Uliassutai, and the other southwest to the province of Sas- 
saktu Khan. The northern trail they said soon branched off to the Nari- 
banchi monastery, about forty miles away. There lives a famous lama hyp- 
notist who was well-known to the Mongols at oiu- halting place. Ossendowski, 
1922, refers to this "Lama Avenger" in his book, "Beasts, Men and Gods." 

We followed the southwestern trail for some miles and stopped at a dry 
salt lake, with a mountain barrier in front of us. Two much-frightened Mon- 
gols who were gathering salt told us that they were the Sassaktu Khan Moun- 
tains and that the principal monastery of the kingdom lay only a short 
distance beyond them. 


By this time we were almost south of Uliassutai and had not seen sedi- 
mentary exposures of any kind, although we had crossed several basins. The 
chances of finding fossiliferous deposits farther west were so poor that we 
decided to return and not attempt to traverse the mountains. At best, such 
a crossing was certain to be difficult and little was to be gained. We returned 
to camp at Kholobolchi Nor on July 13. 


A party went to investigate a small area of red sediments between Baga 
Bogdo and Ikhe Bogdo. They were gone for two days and found not a scrap 
of bone, although the deposits were splendidly exposed. Morris and Olsen 
made a trip of several hundred miles to the north while we were off in the 
west, and had found no profitable exposures. Our western trip had demon- 
strated that we should have to go beyond the longitude of Uliassutai before 
it would be possible to find basins containing exposed sediments; moreover, 
there was little to indicate that such deposits did exist. We had seen most of 
the territory in the Gobi between the Altai Mountains and the northern 
grasslands. Nothing remained for us but to retrace our steps to Shabarakh 
Usu, leave the main camp there and send an exploring party to find a way 
through the Gurbun Saikhan range. South of those mountains we might dis- 
cover important deposits and new geological strata. At any rate it was a 
region which must be investigated. 


Our departure from Kholobolchi Nor was made July 16, and we reached 
Tsagan Nor for an early tiffin. As I have remarked, the water was gone. In 
place of our beautiful lake, a wide expanse of white mud lay stark and dead. 
A lone sheldrake sat disconsolately in the center of the vast open space, the 
only remnant of the clamoring flocks of wild fowl which had rested on the 
water when we left. We had much amusement with Roberts, who had spent 
so many days with Butler and Robinson in mapping Tsagan Nor. We told 
him that he had mapped a lake which did not exist; we would tolerate no 
"faking" on the Expedition and he would have to throw his map away. 


At Artsa Bogdo we camped at the spring near the trail while we tried to 
hire ponies from the natives. Chaney wished to botanize on the mountain 
while Yoving, Loucks, Butler and Robinson htmted sheep and ibex. Most of 
the ponies near the spring belonged to a near-by yamen, we were told. The 
Biuiat officials keep the Mongols in absolute terror. We had the greatest 
difficvilty in buying sheep, hiring ponies or getting any assistance, although 


these people were old friends. They said that virtually everything they had 
was under the direction of the yamen. 


After starting the men for Artsa Bogdo in one of the trucks, the rest of 
us continued our trip to Shabarakh Usu. Thirty-six miles from that place 
the trail led us over an exposed ridge of red jasper. Lying about upon the 
surface were thousands of partly-worked artifacts and a few completed in- 
struments of chert, chalcedony, agate and red jasper. Many of the specimens 
appeared to be so primitive in their workmanship that we believed them to 
be Mousterian or earlier. We spent some time there, collecting many htm- 
dreds of artifacts, and then went on to Shabarakh Usu, pitching our tents at 
the camp site in the basin. 

The spot never seemed more beautiful than when we returned to it after 
the disappointment of ovu- western trip. Nelson, who had stopped to investi- 
gate several promising archaeological sites with Morris, did not arrive until 
some time later. He entirely disagreed with our assinnption that the artifacts 
we had found along the trail were Mousterian. He said, in his opinion, the 
exposure of the jasper ridge represented the source of supply for implement 
material for the Dune Dwellers of the entire Shabarakh Usu region. It was a 
primitive "stone quarry" where they went to obtain their stone; the crude 
workmanship represented "test pieces" in his opinion, chunks of stone which 
had been thrown away as imsuitable after a few flakes had been removed. 
He believed that the primitive artisans, after selecting the material, returned 
to their permanent camps, such as at Shabarakh Usu, to finish the imple- 

Nelson started a hot controversy in camp. We were not at all willing to 
abandon our idea of a new early Palasolithic cultiu"e without a fight. Berkey 
retiimed with Nelson to reexamine the site and the rest of us marshaled every 
argument we could think of to defend our position. But Nelson set about 
demolishing our theories one by one in the most cold-blooded way. He had 
the advantage of knowing a lot more about the subject than we did. After 
he had arranged a series of specimens in rows, showing the entire method of 
flaking from its beginning to the finished artifact, we had to admit that he 
was right and that the crude specimens were test pieces. 

Such discussions are most valuable. The Expedition's work is so corre- 
lated that every discovery affects the other branches of science directly or 
indirectly. Therefore, when a man advanced a new theory he had to be pre- 
pared to defend it from a half-dozen angles. If it survived the attacks from 
the whole staff he could feel that it was probably correct. It was because of 
these constant discussions in camp that I had no hesitation in making very 


definite statements when the results of our summer's work were annovinced 
each year to the world's press after we had returned to Peking. 


While Olsen and Shackelford remained at the Flaming CliiTs, himting 
fossils, Granger, Berkey, Lovell and I left camp for a week's exploration of 
the little-known cotmtry south of the Altai Moimtains. The Russian map 
showed a great basin directly south of us across the frontier of Inner Mongolia. 
The map is so inaccurate that quite possibly no such basin existed, but the 
region was worth exploring. If it was a low-lying area of sediments, it would 
probably contain fossils. No trails were shown, and, since the IMongols said 
none existed, we prepared for a grueling trip. 

Thirty miles from camp, the Gurbim Saikhan Mountains — the "Three 
Good Ones" — rose to a ragged horizon of low peaks. We could see a deep 
cut, which looked like a pass, and we drove up to it on a beautiful slope cov- 
ered with short grass and wild onions, all in flower. The rocky gateway 
proved to be the entrance to a dry river-bed. It led us in and out among 
rounded hills and over lovely upland meadows to a spring of cold, sweet 
water just over the summit of the pass. Right beside it we saw fine exposures 
of red sediment, but not a fossil bone could we find in an hour's search. Some 
time earlier the geologists had discovered a fragment of dinosaur bone in 
similar sediment to the east, so that they felt sttre that it was Cretaceous. 
The Altai Mountains, they had determined, were a late uplift, which had 
been pushed through the old sediments in Tertiary times. On top of the 
Cretaceous lay a deposit of Pleistocene, also barren of fossils. 

From the southern exit of the pass we looked into a vast basin with ex- 
posures of red sediment on all three sides. But a day and a half of prospecting 
and more than a himdred miles of traveling showed us that it was almost 
barren of fossils. We had a most instructive example of how the country is 
cut and eroded by sudden storms. For an hour the Gurbun Saikhan, ten 
miles away, had been obscured by sheets of falling rain. Suddenly we heard 
a muffled roar and saw a flood of brown water advancing upon us down the 
slope. It came so fast that I had to rtm to keep abreast of it. The chocolate- 
colored flood stripped off a thin layer from the surface of the plain, leaving in 
its wake new ridges and furrows. Thus goes erosion in such a land. 

Though, as the Mongols said, no caravan trails led southward, we made 
three attempts to cross the desert, where the mountain ridges lowered to the 
plain. Twice sand turned us back. On the third day we were successful, and 
for more than a hundred miles we plowed southward over heavy terrain. 
Difficult passes let us through low mountain ranges, and in one we had a 
narrow escape. When we had run rapidly up the smooth slope of a low hill 










VV ■ 


'\ V 



4K MiiiQiw- ■ 




and over the crest, a gorge thirty feet deep suddenly opened in front of us. 
Lovell threw on both brakes, stopping the car six inches from the edge. I 
nearly had heart failure as I looked down into the ravine through the wind- 
shield. We were out of that car in about a second and a half. Then the ques- 
tion was how to get it back from the edge. If the machine slipped over, we 
could walk to camp some two himdred miles away or stay there and die of 
thirst. The brakes were holding, but a good strong puff of wind was all that 
was needed to end our automobile ride. Eventually we worked the car back 
inch by inch after blocking the rear wheels with stones so that it could not 
move forward. That was one of the narrowest escapes of our whole summer 
in Mongolia. Had the car crashed into the ravine, the Expedition would 
have been shy four of its members. 

The coimtry we were crossing was hopeless from our point of view. Nar- 
row, ragged motintain chains paralleled each other east and west; between 
them were sedimentary plains uncut by ravines or gullies, so that there were 
no exposures in which to look for fossils. We pushed steadily southward to 
the edge of a vast area of ragged lava hills swept with yellow sand. From 
the summit of the highest peak we could look forty miles across this sea of 
desolation to the blue ramparts of a mountain chain that rimmed the basin 
we had hoped to reach. Nothing on wheels could cross that sand-drenched 
chaos; a camel might have done it, but a horse would have been ruined in an 
hour. To circle it was out of the question without more gasoline than we had 
to spare. The impossibility of getting through was so obvious that it tem- 
pered our sense of defeat. Had there been a chance, we should have turned 
back more sadly. 

When plowing up to another pass ten miles to the west, v/e discovered a 
solitary yiort tucked behind a mass of rocks. A half-dozen Mongols ran out 
frantically, signaling us to stop. It was a yamen, or official post, on the fron- 
tier of Outer Mongolia. A more useless place for a yamen could hardly be 
imagined for we had not seen a sign of habitation in many miles. 


While we were retviming through a movmtain pass just before dark, two 
great brown animals leaped into view on the saw-tooth rim of the highest 
peak. Lovell saw them first. "Sheep, as I'm alive!" he shouted. There they 
stood, two magnificent rams, silhouetted against the sunset sky. Granger's 
rifle was in the car beside me. As Lovell switched off the power, I fired, send- 
ing my bullet into the quarters of the larger ram. I wonder if any other man 
has ever shot a moxmtain sheep from a motor car! I have killed a good many 
bighorns but never one that did not exact strenuous work. Sheep-hunting 
means hard climbing, skilful stalking, straight shooting. To sit comfortably 


in a touring car and pot a Mongolian argali, the trophy par excellence of a 
sportsman's life, was a new experience. Incidentally, it gives an idea of where 
we had taken that car about as plainly as it could be told. I am not surprised 
that the sheep were too curious to nm away when the roaring black thing 
appeared among their mountain peaks. Our being there seemed so strange 
even to us that at times we could hardly believe it true. 


The discovery of a caravan trail, the Great Mongolian Road, paralleling 
the Altai Mountains, sent us eastward next day for more than a himdred 
miles into a most arid desert. It was a bare gravel floor without even the 
stunted "camel sage" and wild onions which are able to exist where there is 
almost no rain. Carcasses of animals marked the track, telling an eloquent 
story of what a toll of life the desert had exacted from the last caravan that 
passed this way. A short distance from the trail lay the body of a man. What 
had been his story? We wondered if he had lost in the battle with thirst and 
hunger or had yielded his life to disease, alone in the silent spaces of the desert. 

The eastern exploration was as unproductive as that to the south had 
been. Low ridges of IMesozoic igneous rocks and inter -mountain basins of 
undissected sediments formed an uninteresting assemblage to a fossil-hunter. 
Since there was no indication that the country would change for a long dis- 
tance, we returned to camp at the Flaming Cliffs. We saw a few gazelles in 
the desert area and one wild ass, which is the most eastern record for the 
species. The exploration had taken us six himdred miles, and, although we 
were bitterly disappointed in not discovering new fossil beds, the negative 
knowledge was valuable. A vast area had been eliminated from our future 
calculations, and the fact that we had already investigated the most interest- 
ing regions of Outer Mongolia had been determined. 


During our absence, interesting things had occurred in our Shabarakh 
Usu camp. When Young and I visited Urga in late May, we had met a charm- 
ing young Dane by the name of Birck, who was in the employ of an English 
company. This same yoimg Dane had suddenly arrived in camp with a caravan 
of camels. His company had sent him to turn back a herd of ten thousand 
sheep which were on their way to Kweihwating, in North China. The sheep 
were to be diverted to Manchuria; just why, Birck was not sure. He believed 
that war in China was the only possible reason, and, indeed, at the time of 
oxxr departure in the spring, it had seemed certain that Chang Tso-lin and 
Feng Yu-hsiang would fight. If there was war, the sheep would have been a 
heaven-sent food-supply for either of the foes. George Olsen is also a Dane. 


It developed that he and Birck had come from the same Httle town in Den- 
mark; their famihes had Hved only a short distance apart, and Olsen knew 
Birck's father. Of coiurse, we all made the usual remark about how small 
the world is, after all ! 

The war news was somewhat disturbing. If Chang and Feng were really 
fighting, the conflict probably would be near Kalgan or along the Mongolian 
border. In that event, either army would welcome our motor cars with open 
arms when we retiuned. I did not worry unduly about the prospect, since it 
was only July 25 and there was ample time for the war to be over before we 
reached China on September 15. Meantime it was possible that we might 
get more information. Birck had remained at camp only one day; for he 
was obliged to rejoin his caravan, which was plodding eastward toward a 
yamen sixty-five miles off. There he was to await the arrival of his sheep. 


The morning after our return to Shabarakh Usu from the southern trip 
we had an interesting experience. It was the "zero hour" just before day- 
light when consciousness is drowned in heaviest sleep. Suddenly I sat up, 
wide awake, with a strange feeling of unrest vibrating every nerve. Slipping 
out of my fur bag, I stepped through the wide angle of the tent door. In the 
air was a stillness vaguely depressing. A cold nose touched my hand, and 
Wolf, our police dog, whined unhappily. He pressed hard against my legs, 
stretched his head toward the Flaming Cliffs and gave a long howl that 
sovinded to me like the wail of a damned soul. It made me shiver, and I 
buckled a cartridge-belt and a revolver over my pajamas before circling the 
tents with Wolf close beside me. All was quiet. Even the camels, kneeling 
in two long double lines, nose to nose, were sleeping. It was the tomblike 
stillness that was so distiurbing. Back in the tent, I made sure that Walter 
Granger's revolver was in its usual place beside his head. Then I slid into 
my fur bag while Wolf squatted in the door with his nose high, sniffing rest- 
lessly. I did not like it and could not sleep. 

At the end of fifteen minutes, I slowly became conscious that the air 
was vibrating to a continuous even roar, which was getting louder every 
second. Suddenly I understood it all. One of the terrible desert storms was 
on the way! As the first blast bellied in the tent, filling it with a whirl of 
sand, I ducked into the sleeping-bag and pulled the flap tightly over my 
head. A minute later the "wind-devil" had passed and I heard muffled 
ciorses from Norman Lovell. His tent was down, and imder the mass of blue 
cloth I could see a writhing hump. Eventually Lovell emerged, laughing as 
usual. "I'm all right," he shouted, "but this tent looks like the wreck of 
the Hesperus. I'm going back to bed." Dragging his sleeping-bag clear of 


the debris, he crawled in, happy as a marmot in a new hole. The "wind- 
devil" whirled and danced away across the desert and left the air heavy as a 
pall. In the gray light of dawn we could see an ominous bronze cloud hang- 
ing over the rim of the basin to the south. Evidently there was more of the 
storm to come, but, since it might miss us, we decided not to wake the camp. 
Ten minutes later the air shook with a roar louder than the first, and the 
gale struck like the burst of a high-explosive shell. Even with my head cov- 
ered I heard the crash and rip of falling tents. It was impossible to see, but 
I felt for Granger with one foot. He was lying across a green suitcase, his 
face protected by a shirt. As our tent swept away, he had leaped to save 
the box that contained six tiny fossil Cretaceous mammal skulls, the most 
precious treastires of all our collections. For fifteen minutes we could only 
lie and take it. While I was feeling for Granger, the sleeping-bag had been 
torn from imder me and the coat of my pajamas stripped off. The sand and 
gravel lashed my back until it bled, and poor Granger on the mammal skulls 
fared no better. 

Suddenly the gale ceased, leaving a fiat calm. The camp was com- 
pletely wrecked. All of the fifteen tents were down, and men were slowly 
emerging from the debris, swearing good-naturedly in English, Chinese and 
Mongol. Our tent was split from end to end, and its contents were piled in 
the most chaotic mass I have ever seen. A trail of litter showed the path of 
the wind toward the "tamarisks" where the Dune Dwellers lived twenty 
thousand years ago. The "tamarisks" looked like Christmas trees, each one 
bearing fluttering streamers of shirts and trousers and dabs of snowy cotton. 
A half-dozen chairs and folding tables had been smashed. Basins, clothes 
and plates were sucked into the whirling vortex, carried hundreds of feet 
into the air and scattered over the desert for a half-mile. Had our cars not 
been facing the wind, they certainly would have been overturned. Never 
have I known so violent a gale. It must have reached a velocity of one him- 
dred miles an hour. Fortvmately, it had passed in fifteen minutes and there 
was an interval of cahn before we had to face the wind again. Every man 
considered it a joke. While we hunted in the growing light for our belongings, 
without a suggestion from me the cooks made coffee and fried antelope steak. 
In a half-hour breakfast was ready. 


Late in the afternoon of the same day. Nelson and Morris returned from 
an examination of new Dune Dweller sites. Morris was ill with the same 
malady that had affected almost all of us at some time dioring the past month. 
He had violent chills and fever and an intense aching all over the body. Doc- 
tor Loucks said it was more like influenza than anythmg else. 


July 27 was another windy day. Granger and Lovell went out with a 
Mongol who had reported exposures northeast of camp, but they proved to 
be unimportant. The same afternoon, Tserin rode into camp on a pony with 
a message from Young, who had stayed with the party at Artsa Bogdo. He 
wrote that the truck clutch had broken and asked for a new part which we had 
in another car. I sent the Chinese chauffeur, Wang, in one of the cars with 
food and spare parts, and the next day the men arrived. They had killed 
eight ibex, one with forty-one-inch horns. They said that there were vir- 
tually no sheep on the movmtain this year. Formerly we used to see almost 
as many sheep as ibex, but for some unaccotm table reason they had suddenly 
deserted Artsa Bogdo. It was delightful to have the entire Expedition as- 
sembled again. There was much to talk about, for every group had had 
interesting experiences. 


Olsen had made an important discovery the previous morning. In the 
extreme eastern end of the basin he found twelve dinosaur eggs biuied in 
soft sand. They were by far the largest and finest that had been discovered 
and amply justified the slogan, "Bigger and better eggs" for the 1925 Expedi- 
tion. The eggs had dropped out of a low sandstone shelf which was consid- 
erably weathered, and Olsen had only to brush them out of the sand. The 
eggs were elliptical and about nine inches long. In fact, they were nearly the 
shape of a loaf of French bread. Their beautifully striated shells showed a 
variety of patterns on the same egg. Only a short distance from them, Liu 
Hsi-ku discovered another nest of the long, slender, thin-shelled type. Quite 
evidently the second lot was the product of a different species, and probably 
genera, of dinosaurs. The new additions brought our collection of eggs for 
the year up to thirty-five or forty, including the specimens which were some- 
what broken. 



Since our exploration south of the Altai Mountains had not produced 
positive results in the way of new fossil fields, the only alternative was to 
retiu"n to Ula Usu, where we knew there were extensive tinexplored deposits. 
I had left this region as a reserve in case the far western Gobi did not prove 
as interesting as we expected. On August 2 we left Shabarakh Usu with 
much regret. This single spot had given us more than we had dared to hope 
from the entire Gobi Desert. The first dinosaur eggs known to man, a hun- 
dred skulls and skeletons of new dinosaurs, eight Mesozoic mammal skulls 
and the primitive hirnian culture of the Dune Dwellers, all had come from a 
few square miles in this lovely basin. Is it surprising that I was filled with 


regret as I looked for the last time at the Flaming Cliffs, gorgeous in the 
morning sunshine of that brilliant August day? I suppose that I never shall 
see them again! Perhaps some day I may view the cliff's from the window of 
a trans-Gobi train, but my caravan never again will fight its way across the 
long miles of desert to this treasure-house of Mongolian pre-history. Doubt- 
less it will be the hunting-ground of other expeditions for years to come. We 
have but scratched the siuiace, and every season of blasting gales will expose 
more riches hidden in its rocks. Who can tell what will come from a place 
that has already given so much? 



When, a few miles east of Shabarakh Usu, in 1922, the clutch of one of 
the trucks broke while we were crossing a small sand-wash, we named the 
spot "Clutch Camp." In 1925 a like accident happened to a truck in exactly 
the same spot. It delayed us two hours, but we camped at night forty-three 
miles from the Flaming Cliffs and eleven miles from the yamen where the 
young Dane, Birck, was awaiting his sheep. Lovell drove over with one of 
the men to inform the yamen that our camels were coming in a few days and 
to let them pass. Birck returned with Lovell to our camp for the first decent 
dinner that he had had since his last visit with us. He had been living on 
Mongol food ever since leaving Urga. 


Birck spent the night at camp and waved us farewell as our cars roared 
away in the morning. The going was excellent and we made good progress. 
We had been playing with scattered herds of antelope along the trail, waiting 
to pick up two or three yearling bucks for meat, and my car was a mile in 
advance of the fleet. An exciting run had just ended and two fat gazelles 
were slung on the fenders of the car, when Doctor Loucks shouted to stop. 
Half a mile to the north lay a low, ragged mass of rocks, the root of an 
ancient mountain peak. On the very summit, silhouetted against the 
sky, stood two magnificent bighorn sheep quietly gazing at us. The glasses 
showed us that they carried superb horns, great circlets at least fifty inches in 

It was McKenzie Young's turn to get a sheep, so we waited for his car 
to arrive. Meantime, the animals remained motionless as though carved 
from granite. Not tintil we were less than a quarter of a mile away did they 
slip over the crest and disappear. We could not find those particular sheep 
again, but they gave us an idea. Jichi Ola — an elongate, rugged mass of 



sediments — rose abruptly out of the plain fifteen miles to the east. It also 
should have sheep and perhaps we could get them from the car. I had shot 
a fine ram while on the southern trip, and it would be a decided novelty to 
hunt one of the wildest and most difficult of big game animals from the cush- 
ioned seat of an automobile! 

The tents were pitched close up against the base of Jichi Ola. This is 
one of a number of low eminences which continues the trend of the Altai far 
to the east beyond the Gurban Saikhan. Each is a simple and single ridge. 
Each has been much worn down and is girded by a broad rock-shelf about the 
base. Jichi Ola is about eight miles long and three miles wide. It stands 
four hundred feet above the level of the alluvial fans about its base. On all 
sides the arid reaches of the desert undulated like the swelling surface of a 
vast brown sea. A hundred yards behind the cook tent, Shackelford flushed a 
woodcock. A bird of paradise would have been no more out of place than 
this shy inhabitant of wooded swamps out there in the center of the Gobi! 
But we realized that the woodcock was migrating southward and had lost its 
way. Wisely it had chosen to lie concealed among the rocks imtil night came 
and it could resume its journey safely. In 1922 at Ardyn Obo I had found 
one in a similar place. 

Four or five of us hunted the next day, and most of the men got sheep, 
but they were all in the highest peaks. Doctor Loucks returned in the after- 
noon to report that he had killed a wolf and two sheep toward the end of the 
mountain; he thought we could get fairly close with an automobile. Robinson, 
Loucks, and I skirted the base of the ridge in the touring car and almost im- 
mediately saw two sheep standing on one of the higher peaks. They ran 
toward us as we advanced, apparently fascinated. Keeping the engine run- 
ning, we stopped. "Robbie" slipped behind a rock and at four himdred 
yards knocked over a yoimg ram. Evidently the time had come to hunt 
from the car, for as shadows lengthened the sheep worked out from the peaks 
to the lower slopes to feed. By careful manipulation we worked the car 
through rocky gateways far up into valleys which cut deep into the motm- 
tain. The roar of the engine echoed like a machine-gun among the cliffs, but 
it seemed to attract rather than frighten the animals. We saw fifteen, I 
believe, and shot three — not such a bad record for bighorn sheep, especially 
when we sat comfortably in a motor car all the while! I think none of us 
will ever forget the drive back to camp through a narrow defile. Exactly in 
the center of the gateway hung a crescent moon partly eclipsed, which threw 
a wan, -unreal light among the rocks. In the path of our headlights, kangaroo 
rats leaped and danced like elfin sprites, and once the dim shadow of a wolf 
crossed into the darkness of the plain. 

The next day while Granger was visiting a red outcrop which stood 


isolated on the desert four miles from the mountain, he came upon a band of 
female sheep led by a yoimg ram. Following them in his car for a mile or so, 
he found that the highest speed they could reach was twenty-five miles an 
hour. They tired quickly, however, and finally gathered into a compact 
group. Running up to within a hundred yards of them he stopped. The 
sheep looked at him as much as to say: "It's your next move." While he 
was trying to get his camera they decided to leave and the group suddenly 
separated, each one dashing wildly for the nearest point of rocks. 


Our next stop, August 6, was at Golobai-in-Ola at the entrance to a beau- 
tiful rocky canyon. It was a perfect place for a camp, wild and rugged, be- 
side a dry stream-bed piled with huge boulders and dotted with poplar trees. 
We stopped at a yurt a few hundred yards away where there was a lame 
woman about forty years old. The poor woman tried to run when she saw the 
car and then stopped, trembling violently. Utterly terrified, she could hardly 
talk, but we gave her two cigarettes and Tserin assured her that she need 
have no fear. When we drove down to the spring in the car, a young girl 
dived behind a boulder like a rabbit; when she saw that she was discovered 
she leaped for the side of the canyon and ran up the rocks. An hour later 
after the tents were pitched we sent Tserin up to inquire about renting some 
ponies. The yurt was gone. In that short time the inmates had disappeared 
with all their sheep. The only things that remained were the two cigarettes 
which I had given the lame woman, carefully placed on a stone where the 
yurt had been. 

Berkey and Morris fotmd that this was a key point for determination of 
the relation between the granites and graywackes and wished to remain a 
day for a closer stud5^ Everyone was delighted, for it was such a beautiful 
spot that it would have been hard to pass it by. Butler, Robinson and Loucks 
hunted sheep the next day but they saw only ewes. The lack of game may 
have been due to the excessive aridity of this entire region. There was al- 
most no vegetation on the mountain ridges. 


We continued eastward on August 8, and that night camped at a well 
named Gutul Usu. A few miles to the north of Gutul Usu Doctor Berkey 
discovered a deposit of low-grade iron which contained a good deal of chro- 
mivim, at Lat. 43° N., Long. 108° 20' E. He saw about one hundred tons lying 
on the surface three miles north-by-east of the well. Were fuel and transport 
available, it would be a valuable deposit; as it is, the iron is worth little. 
This is the only spot in which we had found minerals of any value in all the 


country we had investigated. North of the desert there is considerable val- 
uable metal, but we did not go into that region. 


On August 9, we reached the yamen where our caravan had been de- 
tained in the spring. It was at the point where we should again cross the 
frontier into Inner Mongolia. We were curious to know what would happen. 
I carried enough documents from the Urga government to paper half a room, 
but in the spring they had been ignored by these Buriat officials. All summer 
we had had with us an officer of the Secret Service, part of whose duty it was 
to see that there was no more difficulty at the yamens and to vouch for the 
Expedition. Of course, he was there principally to report upon our activities, 
but as we had nothing to conceal and he was quite a decent fellow, he did not 
annoy us. 

At the border yamen on August lo, we dropped our Secret Service official 
and his bags. The entire personnel of the yamen had changed since our last 
visit; whether my complaints in Urga were responsible for the change, I do 
not know. The new officials appeared to be of the same type. A hard-faced 
Russian was in charge and an insolent yotmg Buriat surveyed our cars with 
a supercilious smile but made no comment about them. The officials prom- 
ised to let our caravan pass, but that, of course, meant nothing. We told 
the officials that in a few days we would send two cars up to the caravan for 
gasoline, and they promised to let them pass. 


When the yamen had been left behind and we were once more in Inner 
Mongolia we felt like celebrating. We got information that as yet there was 
no fighting in the Kalgan area, but that was all. 

A few days later when Lovell and Roberts appeared at the yamen with 
the trucks, they met the same unbearable insolence. Gtms were drawn on 
both sides, but when the Buriats saw that our men would not be stopped 
they allowed them to go. 

Although the officials had promised faithfully not to interfere with our 
caravan, we took no chances. From Shara Mvunm six of us drove a hundred 
miles back to the yamen. We were all heavily armed and determined to 
enforce the privileges given us by the Urga government. We were a grim- 
looking party when our car roared up to the yamen, and the officials could 
see easily enough that we did not intend to argue with them concerning the 
right for our camels to pass. It was a great relief when the entire Expedition 
was across the border. We camped at Ula Usu for a few days while Berkey 
and Granger made a reconnaissance to the northeast along the west side of 


the Shara Murun valley. They found what they believed were very rich 
fossil deposits. 


The work of the topographers was virtually finished and Shackelford 
could use the time to great advantage in Peking developing films. More- 
over, I wanted to know definitely about the political situation. I had no 
wish to run into a war where our cars and equipment would doubtless be 
confiscated by one of the contending generals. Therefore I decided to go in 
with two cars, leave Butler, Robinson and Shackelford in Peking and retiun 
for another month in the field. Chaney also decided to go. He had done all 
that he could in the palaeobotanical line and he was anxious to complete his 
collections of living plants by going through the grasslands where the autumn 
flowers were then in bloom. 


Doctor Chaney has the following general statements to make about the 
botany of the regions we visited: 

"The botanical collections comprise between four htmdred and fifty and 
five hundred species, of which there are four trees — willow, poplar, birch and 
elm; about thirty bushes and shrubs, two vines and the remainder herbs. 
The most abtindant families are the grass family with more than twenty 
species, and the pea family with about thirty species. 

"On the dry desert plains which comprise the most extensive habitat, 
the characteristic plants are several species of bush pea with yellow and pink 
flowers, the salt bush which is especially common in sandy places, the plains 
onion which is certainly the most abundant plant individually in Mongolia, 
and the camel sage. During the rainy season a large number of herbs come 
up, including thistles, iris, morning glory, vetch and grasses. 

"Bordering the lakes and swampy places, the tamarisk, a taU bush with 
pink bvinches of flowers, is common, together with iris, buttercups, sedges, 
and reeds which may reach a height of eight to ten feet. 

"In the mountains the variety of plants is greatest, with marigolds, pop- 
pies, orchids, forget-me-nots and daisies giving color to whole grassy meadows. 
There are many bushes, including a gooseberry and a bush-honeysuckle which 
is the favorite food of the ibex. In the canyon bottoms cottonwoods and 
willows grow on the flats, and under the rocks there are a few delicate ferns. 
On the rocky slopes the rhubarb, mentioned by Marco Polo in his early 
accovmt, is abundant. 

"The grasslands in Inner Mongolia also have a great variety of plants, 
among the most abundant being mints, daisies, thistles, asters, legumes and 


many grasses. Along the stream courses are low elm trees, which in the dryer 
areas are no more than a foot or two tall when full-grown. 

"The Mongols have few sources of plant food. The rhubarb, the onion, 
of which a giant species grows in the mountains, the dune-berry, and in some 
cases the grain from a tall grass are the only ones known to be eaten. A great 
many plants are used for medicine; the ashes of a small mint called by the 
Mongols temen-sul (camel's tail) are put on bums and wounds; the root of 
Ephedra (Mongol name dzergene) is used for rheumatism; a small mountain 
dandelion (called tzetz-serba — stinging yellow) is a cure for headache and 
fever, the dried flowers being ground and made into tea; and a tall aster, the 
Tibetan name of which is rtida, meaning sour root, has roots which are used 
for stomach trouble. 

"A few plants are used for making baskets and pails, including the willow 
(burgase, meaning flexible) and the cotton wood {toree, meaning hollow). The 
bark of the willow is used for tanning leather. All of the larger bushes are 
used for fuel where available, especially the salt bush and the greasewood, 
which is called cheren-tule — blooming fuel. Various plants have a part in 
religious ceremonies, especially the arlsa, a low juniper growing in the moun- 
tains, which is burned as incense in the temples, and a tall grass called deresen 
which is burned as a sacrifice to evil spirits. 

"Many of the Mongol names for plants have attractive meanings, as 
botiila, a small white mountain flower, the name meaning "the flower that 
makes the hills beautiful"; a slender grass called zormasu, meaning mouse- 
tail; a bush called emegenschylbe or woman's calf, because it is so graceful; 
and a thorny pea called khargana — with needles. 

"Further studies of the distribution of the living plants of Mongolia 
will, it is hoped, throw light on the past climate. At the present time, only 
the trees may be mentioned in this connection. In the deep canyons on the 
side of Ikhe Bogdo there are birch trees growing, and in the canyons and 
washes along much of the Altai, cottonwoods are common. Neither of these 
trees can grow in the surrounding lowlands, nor could their seeds, even though 
winged, be blown the several hundred miles from the nearest forest on the 
Arctic Divide. It may be concluded tentatively that these trees represent 
relics of a forest once continuous or at least widely distributed over Mongolia ; 
reduction of rainfall may be supposed to have destroyed this forest in the Gobi 
proper, leaving only a few survivors in the protected mountain canyons."^ 


A few days before we started for Peking, Shackelford had an opportunity 
to photograph the greatest herd of antelope that I have ever seen. We dis- 

' Chaney, R. W., Personal communication. 


covered them one morning six miles from our Ula Usu camp, streaming up 
to the plains out of a great basin. Thousands upon thousands of bucks, does 
and fawns poured in a yellow flood over the rim and spread out like a vast fan 
upon the upland. Shackelford had his motion-picture camera strapped in 
the back of the car and we worked with the herd for hours. But it was rather 
unsatisfactory because so long as they remained on the flat plain they ap- 
peared in the pictures only as a long line of moving animals. 

It was certain that they would not travel far, for the feed was excellent 
where they were. Therefore, in the morning we went out again as soon as 
the light was strong. This time they had arranged themselves as though 
directed by a stage manager ; perhaps fifty thousand were in the bottom of an 
enormous valley where from the rim we could "shoot" down at them with 
the telephoto lens. There was a light wind and for the first time in my life I 
could smell live antelope. A mile away the squalling of the babies could be 
heard. With the glasses we could see them nursing and playing. All the 
intimate details of domestic antelope life were carried on before onx eyes. 
Sometimes a thousand or so would dash at fiall speed through the center of 
the herd, only to stop abruptly and begin to feed. The mass was in constant 
motion; hardly for a moment was any part of it stationary, although the 
animals were entirely at peace. 

I was surprised not to see wolves in the vicinity. One might think that 
such a vast gathering would have attracted wolves from miles around, but as 
a matter of fact wolves are remarkably scarce in Mongolia. After we had 
watched our antelope for nearly an hour and exposed a thousand feet of film, 
we dashed down the long slope directly into the herd. We were almost on 
them before they decided that it was time to really run. Then it was most 
amusing to see them leap over each other to avoid the car; with ears laid 
back the babies put every ounce of strength into the race and for a mile or 
two they could do fully as well as their parents. The herd divided into many 
units and we chased one after another tintil the plain was alive with antelope 
running wildly about in search of their husbands, wives or children. But 
within a few hours they had again collected into a compact mass, and in 
the afternoon we saw them from afar as a splash of yellow on a vast green 

As I have remarked, it is only the grassland species, Procapra gutturosa, 
that gathers into such vast herds. In the spring just before the yoimg are 
bom, the females collect on a flat plain and separate as they drop their young. 
In the fall, bucks, does and fawns again assemble. The long-tailed desert 
species, Gazella stib gutturosa hillieriana, never herd. Doubtless that is be- 
cause no spot on the desert where they live could give sufficient feed for more 
than a hundred or so. 



We left Ula Usu for Kalgan with two motor cars just after a heavy rain, 
and in the afternoon had an experience which might easily have cost us a 
car. A dry river-bed barred our way. Butler and Chaney prospected it and 
waved me to come on. Forttmately, there was a fairly steep bank on the 
other side and I started across at forty miles an hour. Suddenly, "plop!" 
I had a sickening sense of everything going out from under me as the car 
dropped into a quicksand well. It was the same type of death trap into which 
the Bahichitherium, whose legs we found, had sunk some millions of years 
ago. Had we not had another motor with us, a million years from now some- 
one might have excavated a fossilized Dodge Brothers car in just the same 
way. The quicksand was narrow and the speed of the car had carried its front 
wheels across the well. The rear end was sunk at a dangerous angle; with the 
"pull out" cables on the other motor car we drew it to firm ground. 

There had been an imusual amount of rain during the summer and the 
grasslands were blazing with flowers. Chaney reaped a harvest of new spe- 
cies, for we had arrived at just the right time ; a week later many of the flowers 
had withered. As we neared Kalgan we began to get bits of information re- 
garding the political situation in China. There was no war, but that always 
disturbing element, the Chinese student, had engineered a strike and a boy- 
cott on British business firms. It had been going on all summer; that was 
why Birck had had orders to intercept his sheep and divert them to Man- 
churia. The news was comforting so far as we were concerned, for evidently 
it was quite safe to proceed to Kalgan. 

Peking was seething with excitement; therefore it was enjoying itself 
hugel3\ It seemed very strange to come into the flower-filled courtyards of 
my house and don formal evening dress within a few hours of leaving Mon- 
golia. But I had given myself only three days in which to enjoy civilization. 
As soon as a few necessities had been collected and I saw that the needs of 
the men who had returned were attended to, Yoimg and I returned to Kalgan. 


Some intimate friends of mine, the Danish Minister, Mr. H. de Kauf- 
mann, and Mr. and Mrs. Mason Sears of the American Legation, were con- 
templating a trip to Urga, so I invited them to visit our camp near Baron 
Sog, en route. Mr. Robert Williams took them in his car. 

Floods of rain had converted the grasslands into veritable bogs and we 
had rather a messy time of it on the return journey; instead of two days, it 
took us four to reach camp. The Shara Munm, in place of the usual dry 
stream-bed, had become a raging torrent. With the greatest difficulty we 
managed to negotiate a crossing. In doing so we were cheered by the sight 


of our blue tents almost opposite to us on the high west bank of the valley. 
The camp was pitched on the very rim of the basin, four miles north of the 
Baron Sog monastery. 


After we had eaten tiffin, Granger took us out to the fossil fields. He 
and Berkey had reached the conclusion that they were dealing with a new 
geological horizon, probably Lower Oligocene. George Olsen had exposed 
several specimens, but the most interesting was the skull of an extraordinary 
beast that Doctor Loucks had discovered. We had not the remotest idea 
what animal it represented. In fact it was not described until after the 
1928 Expedition, when others of the same genus were discovered across the 
valley. It proved to be a titanothere, representing a new phylum. Professor 
Osbom, 1929, named the specimen Embolotherium loucksi. I wiU describe 
this remarkable beast in more detail in a later chapter. 

The specimen was in such bad condition that one with less experience 
and patience than Walter Granger would have been imable to remove it at 
all. The bone literally was in powder and could be blown away. Granger 
soaked it first with gum arable, which cemented the minute particles to- 
gether; then he stippled on Japanese rice-paper, and when this was dried 
he was able to expose a little more of the bone and repeat the operation. 
Eventually it was bandaged with strips of burlap soaked in flour paste which 
formed a hard shell. 

The morning after our return to camp, Granger came in to report that 
Chth, one of the Chinese collectors, had discovered an enormous skull. We 
all went down to watch the excavation, for that is the most interesting part 
of fossil collecting. Just the tip of a great bone was exposed, and as Granger 
worked away the surrounding matrix, it proved to be the occipital part of a 
skidl. It was so large that at first we supposed it to be another Baluchitherium, 
but as the work progressed it became evident that what we had was a titano- 
there. The frontal region was gone so that we could not see whether it had 
carried a great forked horn as did the American Oligocene titanotheres, but 
the teeth were excellently preserved. 


Meanwhile Nelson had found a rich field in the same area which pro- 
duced the fossils. On a gravel slope facing west were twenty or thirty piles 
of rock which indicated himian work. They were in orderly arrangement 
and he was convinced that they must represent burials. It required con- 
siderable effort to remove the rocks, for some of them were huge slabs sunk 
several feet into the earth. Two graves were empty, but one produced inter- 


esting results. First, he encountered heavy timbers beautifully preserved; 
under these lay the perfect skeleton of a man. He must have been five feet 
ten or eleven inches tall. Beside him lay a birch bark quiver filled with ar- 
rows. Some of the shafts were of wood ; others were partly of reed tipped with 
wood. The points were iron but the metal was badly corroded and in poor 
condition. The bow had separated into half a dozen pieces. 

To me the most interesting thing in the grave was a saddle, upon which 
the man's head was resting. He must have worn a turban, for bits of the 
cloth still adhered to the skull. The saddle was well-preserved, and when 
Nelson brought it to camp it proved to be a perfect McClellan type such as 
our army uses to-day. We had several with us, and the similarity was amaz- 
ing. General McClellan without doubt thought that he had developed a new 
saddle, just as we supposed that we were the original discoverers of the dino- 
saur eggs. But in both cases primitive dwellers of Mongolia had made the 
discoveries centuries before we were bom. The saddle is quite unlike that 
used by Mongols or Chinese to-day or in the past, so far as I am aware. 

Nelson thought that the grave must be at least a thousand years old 
and possibly more than that. The fact that it was placed in a well-drained 
slope, in an extremely dry desert, undoubtedly accounted for the splendid 
preservation of the wood and bones. It was impossible to identify the skele- 
ton in the field, but its racial characters can be determined by study at the 


Berkey and Morris fovind evidences of other prehistoric people in rather 
an interesting way. They were sitting in a Mongol's yurt, twenty miles from 
Ula Usu, when Berkey's attention was caught by a small nugget of copper 
ore lying on the family altar. The Mongol was definite in his information that 
it came from a spot quite near a temple fifteen or twenty miles to the south. 
The geologists visited the place and found evidences of surface mining opera- 
tions on a large scale. A vast pit had been excavated in the hillside; so large, 
indeed, that at first they did not believe it could be artificial. The copper 
was not in veins, and the deposit had been so thoroughly worked that but 
comparatively little remained. They studied the place carefully and came 
to the conclusion that the mining operations must have ceased at least a 
thousand years ago. 


While we were at the camp where Nelson had found the grave, the ante- 
lope herd which Shackelford had discovered paid us a visit. We heard them 
pouring down into the basin during the night, and two days later the whole 








mass came up again not more than four hundred yards from camp. While 
we were at breakfast the bleating of the fawns and the "tap tap" of thou- 
sands of tiny hoofs brought us all out of the tent. A vast yellow blanket of 
moving forms was flowing over the edge of the bluff onto the plain. Wolf, 
my police dog, went wild with excitement. He chased group after group 
until he was exhausted, but the antelope could leave him behind so easily 
that they bothered to run only when he was almost upon them. 


When we were ready to move to another fossil locality, we asked the 
priests of the Baron Sog monastery for permission to deposit our specimens 
in their care until the caravan arrived. We had done this half a dozen times 
during the preceding two years. They said that we could leave gasoline or 
rocks, but no fossils, because last summer many horses and sheep had died 
in the vicinity and that doubtless it was due to the bad influence of the 
"dragon bones." 

Our next camp, ten miles to the north of Baron Sog-in-Sumu, was very 
similar to the one we had left. The tents were pitched on a great promontory 
which projected far out into the basin. Near them was an obo, or religious 
monument, and shortly after our arrival two lamas came to call. They were 
delegates from a temple, Tukhum-in-Sumu, four miles away, and asked us 
to be particularly careful not to shoot or kill any birds or animals on the 
bluff. It was a very sacred spot and the spirits would be angry if we took 
life in the vicinity. Of course, I agreed to respect their wishes and gave orders 
at once. But we had promised more than we could fulfill, as events proved. 


In the first two hours of prospecting, three pit-vipers, Agkistrodon, were 
discovered close to the tents. A few days later the temperature suddenly 
dropped in the late afternoon and the camp had a busy night. The tents 
were invaded by an army of vipers which sought warmth and shelter. Lovell 
was lying in bed when he saw a wriggling form cross the triangular patch of 
moonlight in his tent door. He was about to get up to kill the snake when 
he decided to have a look about before he put his bare feet upon the ground. 
Reaching for his electric flashlamp, he leaned out of bed and discovered a 
viper coiled about each of the legs of his camp cot. A collector's pickax was 
within reach and with it Lovell disposed of the two snakes which had hoped 
to share his bed. Then he began a still-htmt for the viper that had first 
crossed the patch of moonlight in the door and which he knew was some- 
where in the tent. He was hardly out of bed when an enormous serpent 
crawled out from under a gasoline box near the head of his cot. 


Lovell was having rather a lively evening of it — but he was not alone. 
Morris killed five vipers in his tent, and Wang, one of the Chinese chauffeurs, 
found a snake coiled up in his shoe. Having killed it, he picked up his soft 
cap which was lying on the ground and a viper fell out of that. Doctor Loucks 
actually put his hand on one which was lying on a pile of shotgun cases. We 
named the place "Viper Camp," because forty-seven snakes were killed in 
the tents. Fortunately, the cold had made them sluggish and they did not 
strike quickly. Wolf, the police dog, was the only one of our party to be 
bitten. He was struck in the leg by a very small snake, but since George 
Olsen treated the wound at once, he did not die. The poor animal was very 
ill and suffered great pain, but recovered in thirty-six hours. 

The snake business got on our nerves and ever>'one became pretty 
jumpy. The Chinese and Mongols deserted their tents, sleeping in the cars 
and on camel boxes. The rest of us never moved after dark without a flash- 
light in one hand and a pickax in the other. When I walked out of the tent 
one evening, I stepped upon something soft and round. My yell brought 
the whole camp out, only to find that the snake was a coil of rope ! We had 
to break my promise to the lamas and kill the vipers, but our Mongols re- 
mained firm. It was amusing to see one of them shooing a snake out of his 
tent with a piece of cloth to a place where the Chinese could kill it. The 
vipers were about the size of oiu" copperheads, or perhaps a little larger. 
While their fangs probably do not carry enough poison to kill a healthy man, 
it would make him very ill. 

The snakes inhabit bluffs throughout the desert, like the one on which 
we were camped. Their great number at this particular spot was due to the 
fact that it was a sacred place and the Mongols would not kill them there. 
This viper appears to be the only poisonous snake in the Gobi, and we col- 
lected but one non-poisonous species. The climate is too dry and cold to 
favor reptilian life. 


The new camp proved to be just as rich in fossils as it was in snakes. 
One place which evidently had been the bed of a stream that had flowed 
there during Eocene times was a veritable quarry of fossil bones. Twenty- 
seven jaws were exposed at one time in the same layer, and it was necessary 
only to scrape off a few inches of sediment in almost any spot to imcover 
valuable specimens. We got skulls of a peculiar beast known as the Chali- 
cothere, a veritable paradox. It is a "clawed-hoofed animal." The head 
and neck are like those of a horse, the teeth like a rhinoceros, and the feet 
like nothing else on earth. Instead of hoofs the creature was armed with 
claws. Why such an anomaly was developed no one can tell. There must 


have been some good reason, for nature does not produce such extraordinary 
appendages haphazardly, but thus far the explanation is obscure. 

The region must have swarmed with a little hoofed beast known as Lo- 
phiodon, for the palseontologists obtained a great collection of jaws and skulls 
which represent many vmknown species and genera. We found no trace of 
horses in the very old formations. This is a great surprise, for the unknown 
five-toed ancestor of the horses is one of the types which we confidently ex- 
pected to discover. Four-toed horses are present in the Eocene of both 
America and Europe, and we are certain that the ancestral stock developed 
in Asia, but as yet it has eluded us. Nevertheless it must be there, although 
I have come to believe that it will be foimd in southern Siberia rather than 
in Mongolia. 


Berkey, Lovell and I drove north in the Shara Murun valley for sixty 
miles to the border of Outer Mongolia. There we were stopped by yamen 
officials with the usual insolence. They would not recognize our passports 
as valid, although we had just come across the frontier. Doubtless we could 
have made them let us pass by showing force, but it did not seem to be worth 
while and we turned back. Off to the east, on the opposite side of the valley, 
we had dimly seen red exposiu-es which in 1928 proved to be rich fossil 

While work was proceeding at Viper Camp, six of us made a five-hundred- 
mile exploration south and west. It was an attempt to find a trail which 
would take us out to Chinese Turkestan through a little-known region of 
Inner Mongolia. We hoped to explore that part of Mongolia during the 
next expedition. We discovered a trail, but since we followed it in 1928 I 
will reserve my description of it until the narration of that Expedition. 


When we returned from the southern trip the caravan was awaiting us 
at the Baron Sog monastery. Rain and light snow warned us that it was 
time to leave if we were not to be caught in the bad weather of early winter. 
The sand-grouse were flocking and golden plover had arrived in thousands 
from the Siberian tiuidras. On September 12, we drove down the slope to 
the basin floor, leaving Viper Camp to the snakes and vultures. Another 
season had ended and we were well content. 


After the 1925 Expedition, Professors Berkey and Morris studied the 
Cretaceous period in Asia as a whole and suggested the following inferences 


as to conditions during this closing period of the Age of Reptiles, approxi- 
mately lOO million years ago: 

"i. The continent of Asia in Cretaceous time was relatively low lying, 
and was bounded on the north, south and southwest by shallow epicon- 
tinental and mediterranean seas. 

"2. There may have been in Cretaceous time a series of near shore ele- 
vations, ranges that foreshadowed the moimtains of to-day, and that occu- 
pied the places where the modem ranges stand. The erosion of these ranges 
would yield sediment to the sea on the one hand, and to the inland basins on 
the other; but no deposits would ever form along their site. This inference 
seems to fit in well with the observed facts. 

"3. The primitive trachydont dinosaurs of Asia may have given more 
advanced descendants to Europe and even to America. The highly special- 
ized ceratopsians of America may have been derived from the primitive 
Asiatic forms, though it is not necessary to derive them directly from the 
Djadochta species. The presence of large sauropods also suggests intercourse 
between Asia and other lands, especially America. But the isolation of the 
faunas in Oshih-Ondai Sair and Djadochta time must be due to barriers of 
some sort. The barriers may have been seas, such as Borissiak and Grabau 
have postulated along the western Siberian lowland; or deserts, giving areas 
of uncongenial environment. Mountain barriers and isothermal barriers are 
unlikely, because no mountains that are likely to have existed in that period 
could have been so uniformly long and high as to prevent migration, espe- 
cially of upland faunas such as the Djadochta types. The Cretaceous was a 
period of mild, equable climate even in the far north, so that frigid zones did 
not shut in the faimas as they do to-day. Whatever barriers kept American 
and European hordes from invading Asia, and for a full period withheld Asia's 
evolving dinosaiirs from Europe and America, probably were either water 
bodies or deserts which could not be crossed."' 

1 Berkey, Charles P., and Morris, Frederick K. (MS. for Vol. Ill, this series, in preparation). 



Shortly after returning to Peking from Mongolia in 1925, Berkey, 
Morris and Roberts left for America. Shackelford soon followed with twenty 
thousand feet of motion-picture film which he had developed in oior labora- 
tory in Peking. Lieutenants Butler and Robinson retiuned to their respec- 
tive military duties, and Doctor Loucks resumed his work at the Peking 
Union Medical College. Granger and Nelson prepared for a winter along the 
Yangtze River, the former to continue his exploration of the fossil pits of 
Yen-ching-kou, near Wanhsien, Szechwan, the latter to examine the numerous 
caves along the river banks in the hope of finding evidences of palaeolithic 
man. It was imperative that I should return to America to obtain additional 
financial support for the Expedition and to stimulate public interest by lec- 
tures and writing. Affairs in Peking were left in charge of J. McKenzie 
Young, assisted by Norman Lovell. 


Doctor Chancy went northward into Manchuria to examine reported de- 
posits of plant fossils. Before he left America he had told me of his hope to 
find the remains of redwood forests in eastern Asia, and his prediction was 
fulfilled at the Fushim Mine of the South Manchurian Railway Company, 
thirty miles east of Mukden. There, in shales above and below the coal, he 
discovered fossil plant remains, the most abundant of which was the red- 
wood Seqiwia langsdorfii; other species were the alder, oak, maple, sycamore, 
poplar and fern. This fossil assemblage is almost identical with the Bridge 
Creek flora of Oregon, which is referred to the Upper Oligocene. Both of 
these fossil floras are closely similar to the redwood forests now living on the 
coast of California. 

While it was known that redwoods occurred in eastern Asia during Ter- 
tiary time, no indications of the existence of a considerable forest had been 



reported. The presence of the other trees, which are so characteristic of the 
living redwood forests, leaves little doubt that a forest approximating that of 
California formerly was present in Manchuria. Doctor Chancy foimd addi- 
tional evidences at the Pataoho Mine of the Fengtien Mining Administration. 
It is probable that the forest extended continuously to Vladivostok, where 
similar fossils have been found. 

Since the living Sequoia must have at least forty inches of rain a year 
for its best growth and does not thrive in regions of extreme climate or low 
rainfall, it may be inferred that its Oligocene equivalent of Manchuria and 
Siberia required about the same conditions. The rainfall at Fushun to-day 
is less than half the amount necessar>' to maintain a forest of this type. In 
discussing the fossil flora of Mongolia and Manchuria and the climatic con- 
ditions which they indicate, Doctor Chancy remarks: 

"Little is known of the pre-Cretaceous floras of Mongolia except that 
trees grew on the borders of streams and that their stems were transported 
as driftwood and deposited in the sand and mud at the rivers' mouths. In 
the course of subsequent mountain-making movements, the characters of 
these early plants have been largely obliterated. The presence of a varied 
flora in the Permian of eastern Chihli suggests that conditions for plant life 
may have been more favorable there than in Mongolia during the Palaeozoic. 

"The Cretaceous flora of Mongolia appears to have been made up of 
good-sized trees, mostly of coniferous types, which occupied the borders of 
lakes together with rushes. There is no record of the vegetation of the up- 
lands where Protoceratops lived; these were doubtless occupied by grass and 
bushes, but the distance to the streams and lakes was too great for them to 
be left in the sedimentary record. From the fact that plant remains are not 
more numerous, the climate during the Cretaceous may be supposed to have 
been fairly dry, so that the trees were limited in number and distribution. 

"The Tertiary flora of Mongolia enters even less into the geologic record, 
suggesting a climate still more arid than that of the Cretaceous. This sug- 
gestion is borne out by the comparatively small size and the twisted char- 
acter of the fossil stems, which in these respects closely resemble the gnarled 
bushes now growing in Mongolia. The almost complete absence of leaf im- 
pressions is a strong indication of aridity, for if there had been many water 
bodies the leaves would have been preserved in the clays instead of disin- 
tegrating in the air as they appear to have done. To the east and south in 
southern Manchuria, climatic conditions appear to have been much more 
favorable for tree growth. During the Oligocene, there was a luxurious 
redwood forest which may have extended continuously to Vladivostok. 

"It appears probable that the great difference in rainfall between Man- 
churia and Mongolia during the Tertiary was due to the range of mountains 


between, which cut off most of the rainfall from Mongolia in much the same 
way as it does to-day. A similar situation is now to be found in California, 
where near the coast the rainfall is high and redwoods and other trees flourish, 
while in the interior, to the east of the Sierra Nevada, trees are almost entirely 
absent. The scarcity of fossil plants in the Tertiary of Mongolia may best be 
interpreted on the basis of such an arid or semi-arid climate, and the evidence 
of the vertebrates is quite in accord with such an interpretation."' 

Doctor Chaney remained in Manchuria for some time and then went to 
the Philippine Islands before returning to America. 


George Olsen fitted out my former large office at headquarters in Peking 
as a laboratory in which to prepare the fossil collections. Under Olsen's ex- 
pert direction, the five Chinese field collectors began to turn out completely 
restored specimens f-ully as quickly as could have been done in the American 
Musevun of Natural History. Establishing this field laboratory was some- 
what of an experiment, for we were by no means sure that the specimens 
after restoration could be safely shipped to America. I may say, however, 
that our greatest hopes were realized. Not only did the fossils reach New 
York with virtually no breakage, but tlie cost of preparation was about one 
sixth what it would have been in the Museum. Chinese are notably skillful 
with their hands, and by the end of the year our native preparators were 
well nigh as good as those in New York. 


The 1925 Expedition, with a foreign staff of fourteen men and twenty- 
six Chinese and Mongols, proved to be too large. Our mobility was sacrificed 
and it became unwieldy for the country in which we were operating. It re- 
quired too much gasoline to move, and too much food to maintain, the party. 
I had come to the conclusion that a foreign staff of ten was the maximum 
number that we could use effectively. I intended to organize the 1926 Expe- 
dition on that basis. 

Since Doctor Berkey could not leave his work at Columbia University 
for the next year, R. H. Beckwith, one of his students and an Oxford Rhodes 
scholar, was substituted as geologist. It was impossible for Major Roberts 
to return, and he was replaced by Captain W. P. T. Hill, U. S. Marine Corps, 
who was detailed to the Expedition through the kindness of Major-General 
Le Jeime, and of the Honorable Curtis D. Wilbur, Secretary of the Navy. 
Captain Hill had had much experience in Alaska and other parts of the world, 
and combined a considerable knowledge of geology with expert training as a 

' Chaney, R. W., Personal communication. 


topographer. Dr. W. D. Matthew had long wished to study the MongoHan 
fossil fields, and we all were delighted when Museum affairs were so arranged 
that he could join the Expedition. 


Just after I had sailed for America in October, 1925, the ship's radio 
picked up news of a civil war which had started near Shanghai. It caused 
me no worry because Shanghai is a long way from Peking and I did not be- 
lieve that the war would spread northward. Even if it did, the northern 
Chinese never fought in the winter. I thought that it would last for a few 
weeks and then be settled in the usual Chinese manner, without much blood- 
shed. But it was just that year when things began to change in China and 
tradition and "good form" in the conduct of wars were completely smashed. 
The trouble did spread to the north like a flame and lasted all winter. Some 
of the bitterest fighting was in December and January near Tientsin. The 
second breach of "good form" was that they killed a good many people. 
Thousands of wounded poured into the city and the countryside was strewn 
with dead. The railroad transport of all north China was paralyzed; for 
weeks no trains ran between Peking and Tientsin, although, according to the 
1900 Protocol, the foreign powers have the right to maintain communica- 
tions from the capital to the sea. Food was expensive and difficult to obtain. 
It was not a question of how soon a thing could be done, but whether it could 
be done at all. Therefore McKenzie Young had his hands more than full with 
preparations for the spring expedition. 

Just about the time he had everything ready, renewed hostilities began. 
Gradually the "People's Army" of the so-called "Christian General," Feng 
Yu-hsiang, was pushed back toward Peking. The only communication with 
Tientsin was by motors which were allowed to pass through the lines when 
there was no fighting actually on the road. 


To buy tons of rice and flour, hundreds of pounds of sugar, coft'ee, beans 
and other food supplies in Peking was impossible. They had to be purchased 
in Tientsin, but that port is eighty miles from the capital, and no trains were 
running. Foiu* thousand gallons of gasoline had been specially packed for 
us by the Standard Oil Company but it, with the food, was all in Tientsin. 

Since our caravan must start in Ivlarch at the latest. Young decided to 
bring the gas and supplies up to Peking by motor. It was a laborious and 
very expensive method, but there was no other way. For three weeks Nor- 
man Lovell made a round trip every day. Leaving Tientsin early in the 
morning, he drove the eighty miles over what is only by courtesy called a 


motor road, and arrived in Peking at noon. While he was eating Itmcheon 
the car was unloaded and he was off again on the retxim trip to Tientsin 
before one o'clock. 

During these days of transport he had many exciting experiences. The 
road swarmed with soldiers of one side or the other, and a more annoying 
collection of uniformed brigands would be difficult to find anywhere. He 
carried a large American flag on the car, but this did not prevent his being 
fired on time after time. There was a "no man's land" of ten or twelve miles 
between the opposing lines, and entering or leaving this area usually was 
attended with a good deal of danger. The advance- and rear-guards in the 
trenches not infrequently took a flying shot at the car just to see if they could 
hit it. Sometimes there was genuine suspicion that it might contain 
a machine-gun party of the opposite side, and the treatment Lovell received 
was distinctly unpleasant. Of course, he would be stopped and questioned 
every few miles, but cigarettes and a plentiful supply of calling cards usually 
were sufficient to get him through. 

Just before I reached China in March, 1926, four new cars for the Ex- 
pedition arrived from Detroit, for our fleet was to consist of eight Dodge 
Brothers cars. Lovell with three Chinese chauffeurs drove them to Peking. 
Unfortunately, they happened to start on the day that the so-called "People's 
Army" were retreating along the motor road. Thousands of troops and lines 
of carts blocked the way. Our cars were heavily loaded with gasoline, but 
the soldiers literally took possession of them. So many men piled upon each 
car that they could hardly pioll even in low gear. The springs were abso- 
lutely flat and had they not been designed for especially rough work they 
certainly must have been ruined. The soldiers were very ugly, and the fact 
that the cars bore American flags and that the road was a recognized avenue 
of traffic for automobiles made not the slightest difference. Lovell's feat in 
getting our gasoline and supplies to Peking is, in itself, sufficient testimony to 
his coiorage and tact. 


It was the first week in March before our equipment was assembled at 
the Expedition headquarters in Peking. The caravan should have left with 
the equipment by mid-February. When the first war rumors developed in 
the early winter. Young had sent word to Merin, the Mongol caravan leader, 
to take our camels far out into the desert where they would be safe. Feng 
Yu-hsiang was confiscating every cart, mule, horse and camel within a hun- 
dred mile radius of Kalgan, and the American flag which floated over Merin 's 
tent would not have protected oiu" camels from the soldiers of the "Christian 
General" or any others. 


Meantime all the railroad cars on the Peking-Suiyuan Railroad had been 
seized by the military. It was virtually impossible to move a pound of freight 
to Kalgan even by paying the most exorbitant "squeeze" for the use of a 
freight car. Sometimes a merchant did manage to get one started from 
Peking, but it was usually sidetracked at Nank'ou, half way to Kalgan, there 
to remain indefinitely. Young had tried by every means to get our supplies 
to Kalgan, without success. 


I arrived in Tientsin on March 27, 1926, with J. B. Shackelford, the 
Expedition photographer. There had been sharp fighting not far from Peking, 
and the authorities were loath to allow Young to drive to Tientsin to bring 
us up. His trip down was comparatively uneventful for those exciting days, 
although he was halted and questioned a dozen times by soldiers of both 
sides. When we returned two days later, the road had been mined in fourteen 
places by. Chang Tso-lin, as preparation for a counter attack by the "People's 
Army." There were thousands of soldiers, but we had no trouble, although 
one became a bit jumpy when driving over the mines which we hoped were 
buried deeply enough not to explode from the impact of our motor. 


A week after our arrival in Peking, I managed to get two freight cars 
for Kalgan through the influence of Mr. C. S. Liu, Director-General of Rail- 
ways. He sent word at four o'clock in the afternoon that they were available 
but said that we must load them at once or they would be taken by the sol- 
diers. We worked a good part of the night taking over the gas and supplies. 
One was a covered steel car and loopholed and bullet-marked; the other was 
a flat car upon which we drove two motors and piled them about with cases. 
Yotmg and Shackelford chaperoned the things to Kalgan without serious inci- 
dent and returned to Peking. Word had been sent to Reverend Joel Eriks- 
son, a Swedish missionary, asking him to bring in the camels, and as soon as 
they arrived I expected to go up with Yoimg to see them started. 


In the meantime Peking had been having daily visitations from an aero- 
plane which flew over the city every morning at ten o'clock and dropped 
eight or ten bombs. The plane stayed so high that its missiles were badly 
directed and did comparatively little damage. A few pedestrians or mer- 
chants were killed at every raid, but the aviator seemed unable to hit the 
barracks or railway stations which were his objective. Fortunately the 
bombs used were loaded with black powder and weighed only eight or ten 


pounds; had they contained high explosives it woiild have been a good deal 
more serious. 

After the first two or three days, Peking accepted the aeroplane visita- 
tions as just another of the nuisances caused by the stupid war, and char- 
acteristically proceeded to derive as much amusement as possible out of the 
experience. "Bombing breakfasts" became a new form of social entertain- 
ment. We used to gather at nine o'clock at the Peking Hotel, breakfast 
comfortably, and then adjoiun to the roof at five minutes to ten. Promptly 
on the hour the plane would appear out of the sky to the south. After a 
preliminary circle we could hear the z-z-z-boom of a falling bomb, quickly 
followed by two others, for they invariably came in threes. Great clouds of 
yellow dust accompanied the explosions. The Chien Men railway station 
near the American Legation was a favorite objective of the airman, and the 
Legation Staff began to get a bit nervous. It was quite obvious that the 
bomber was trying to avoid the Legation Quarter, but his aim was so bad 
that an "egg" might land in the compound at any time. 

On Monday morning, April 12, the plane appeared at nine o'clock, an 
hour earlier than usual. Flying over the South City, it dropped one bomb 
in a Chinese restaurant, killing five men and wounding three; the second 
landed in a primary school and killed several children; the third exploded in 
the middle of a street but did no damage. We supposed that the raid was 
finished for the day. I was due at the Hsichihmen railroad station at eleven 
o'clock to interview the Inspector regarding freight cars for our remaining 
six motors. Lo, one of my Chinese "boys," went with me. 

Just as we were about to emerge into the broad esplanade fronting the 
station we saw people running in all directions. Lo shouted that an aero- 
plane was overhead, and the next instant there was a terrific explosion about 
thirty yards to the right of us. Fortunately the bomb had landed on the 
other side of a high mud wall which protected us from the iron fragments. 
I put on full speed, thinking to get into the depot for protection, but a second 
bomb exploded in front of the car and indicated that the airman was going 
in the same direction. We stopped abruptly. I leaped out of one door of 
the motor and Lo from the other. He rushed down the street and backed 
up against a wall. I dashed for some loaded freight cars standing on the 
track. An engine was just backing up to the cars and it was amusing to see 
the crew fly out of the cab and run in various directions at the crash of the 
second bomb. 

With a dozen Chinese I crawled under the train. As the cars were loaded 
with bags of grain they gave fair protection. My car had solid steel wheels 
and I stretched out between them parallel with the axle. I was hardly in 
position when a bomb exploded with a deafening roar not fifteen feet from 


my shelter. The steel fragments "pinged" against the car wheels like rain 
and I never realized how small I could make myself imtil that moment. A 
few seconds later two other bombs crashed almost simultaneously on the 
other side of the train. One steel fragment came in at an angle and biiried 
itself within two inches of my face. I dug it out and burned my fingers nicely 
for it was red hot. After a few minutes all seemed quiet and I crawled out 
from under the train, thinking that the plane had gone. Instead it had only 
done a circle and was directly above my head. Before I could duck back 
under the train a bomb exploded a few feet away, directly in front of a Chi- 
nese woman. It blew her head off as neatly as though it had been severed 
with a knife. Another bomb landed in front of the wall where Lo had been 
standing and killed four men. Lo escaped only because he had at last obeyed 
my calls to come under the car. 

The wretched airman seemed to have a special hate against that par- 
ticular train and made three attempts to wreck it, dropping fourteen bombs 
within a few feet of it before he disappeared to the eastward. We were hav- 
ing rather a lively time, what with the noise and smoke and groans of the 
people who had been injured! 

The Standard Oil Company own a tank which contained twenty-four 
thousand gallons of oil within a few yards of where half a dozen bombs ex- 
ploded. Had one made a direct hit on the tank there would have been a most 
beautiful conflagration. Lo and I returned to the motor thinking we could 
finish our business with the Inspector, but it was useless. While we were in 
the station the airman returned and peppered us again. We decided then 
that it was an unhealthy locality and made a dash for safety. I was inter- 
ested in the unconcerned attitude of the Chinese. An air raid is a nerve- 
racking experience for me, but the Chinese seemed to take it in the most 
casual way. At other times I have seen them thrown into a panic by a com- 
paratively trivial thing. 


On April 13 there was heavy gun-fire just outside of Peking, and after a 
dinner at the American Minister's, we adjourned to the roof of the Peking 
Hotel. The machine-guns showed in a steady stream of light along the 
southern horizon punctuated by the wide flashes of heavy guns. Hugh 
Bradley, of the Chinese Maritime Customs, who was leaving for America 
with his wife and child, tried to get through to Tientsin by motor, but shells 
were falling uncomfortably close and rifle-fire across the road drove them 
back. The American MiUtary Attache told me that the "People's Army" 
were noticeably stiffening and that they might even push the Fengtien troops 
back to Tientsin. But the usual thing happened. One of the generals of the 


"People's Army" turned traitor, deserted to Chang Tso-lin with thirty thou- 
sand men, and the advance became a retreat. 

I wanted to get through to Tientsin and made a try for it the next day 
with three members of the Expedition — Shackelford, Hill and Beckwith. 
We thought that a large American flag on the car would protect us as it had 
done in former years. The gates of Peking were heavily guarded, but the 
soldiers let us pass. Carts were already coming into the city loaded with 
grain, camp-gear and soldiers. Cavalry streamed by and then thousands 
upon thousands of infantry. They were retiring in good order and seemed 
most cheerful. An officer told me that Chang Tso-lin's troops had taken 
Tungchow, fourteen miles from Peking, and were looting the city but that 
no fighting was taking place. 

We drove on slowly, and eventually passed beyond the rear of the re- 
treating army. For three or four miles the countryside was deserted, houses 
closed and all as quiet as the grave. We were five or six hundred yards from 
the ancient marble bridge at Tungchow when there came the sharp crack of 
a rifle and a bullet struck beside the front wheel. A second later a mass of 
soldiers appeared on the road and bullets began spattering arotmd us like 
hailstones. They had opened fire with a machine-gun, but it was aimed too 
low and the bullets were kicking up the dust just in front of us. The soldiers 
could see the American flag plainly enough but that made not the slightest 

Fortimately at this particular spot the road was wide enough for the 
car to be turned and I swung it about in record time. The bullets now were 
buzzing like a swarm of bees just above our heads. Forty yards down the 
road a sharp curve took us out of sight of the machine-gun. The other men 
crouched in the bottom of the car. Since I was driving I could see all the fun. 
It was a pretty rough road, but the speedometer showed fifty miles an hour 
as we went back. The ride became an exciting one. All the houses which 
had seemed so peaceful actually were occupied by the advance-guards of 
Fengtien soldiers. They had let us pass because of the American flag, but 
when they heard the firing in our rear and saw us returning at such a mad 
speed, they evidently thought that we were anybody's game. Each and 
every one decided to take a shot at us. 

For three miles we ran the gatmtlet of firing from both sides of the road. 
I would see a soldier standing ready with his rifle at the side waiting imtil 
we came opposite. Then "bang" he would let us have it. Sometimes they 
fired in squads; sometimes singly. The only reason why we were not riddled 
with bullets is because the Chinese soldier is the world's worst shot. Most 
of them aimed directly at the car, when they aimed at all, and the bullets 
struck just behind us. Every now and then one would zip in close to my head 


but no one was hit. I really had the best of it because the others could not 
see what was going on and driving the car kept me busy. I expected every 
moment that one of the tires would be hit. A blow-out at that speed would 
have turned us over. 

Before long we could see the rear-guard of the retreating army and the 
sniping at us ceased. Still our troubles were far from being ended. The first 
retreating soldiers, three of them, asked for a ride. I thought that they 
might be a protection and let them stand on the running board of the car. 
Suddenly one of them saw an officer. Without a word he stepped backward 
off the car, rolled on the ground with his right hand under the rear wheel. 
As I put on the brakes it ground his hand and arm into the hard gravel road. 
I have never seen such a sight; his hand was simply shredded. I put on a 
tourniquet to stop the bleeding but he was only anxious for us to go. 

A little farther and we came to masses of infantry. Against my protests 
they piled on the car in such nimibers that it could pull only in first speed. 
The inevitable happened when one fell off, breaking his leg. Things looked 
pretty bad. Three or four of the soldiers worked themselves into a rage, 
cocked their rifles and were just about to shoot us when an officer appeared. 
Fortunately he coiild speak Mandarin Chinese perfectly (the others talked 
a difficvilt Shantimg dialect) and when I explained what had happened he 
cleared a passage so that we could drive off the road into the fields. With 
much difficulty we got through the gates back to Peking. 


Young went up to Kalgan on the evening of April 13 to see the camels 
started. On April 15, the day after our experience in trying to reach Tientsin, 
Shackelford and I decided to join Yoimg in Kalgan. At eight-thirty in the 
morning when we drove to the Hsichihmen gate we found it closed and sand- 
bagged. The soldiers said that no one could leave or enter the city. We 
made the roimds of the other gates and iovcnd them all heavily barricaded. 
The streets were deserted, the shops closed, and a strange air of preparation 
for a great calamity pervaded the city. If the Fengtien troops forced the 
gates, Peking would certainly be looted. Every Chinese of political impor- 
tance who could find a lodging had fled to the Legation Quarter. Most of 
these same men had been loudly demanding the abolition of extra-territoriality 
and the dismissal of the Legation Guards but a few days earlier; yet at the 
first hint of danger they dashed to the foreign concessions as the only place 
of safety! 

A good many foreigners who had cottages at the race-course, seven miles 
west of the city, were caught outside and had to remain there for several 
days in the midst of roving bands of soldiers. The First Secretary and the 


Counselor of the British Legation tried in vain to get the guards to open the 
gates. At last they found a part of the wall where loose bricks had been 
removed by smugglers. A niimber of Chinese were on top and when the for- 
eigners produced a silver dollar a rope jnagically made its appearance. The 
diplomats were haioled imceremoniously up the wall, losing a certain amount 
of "face" but at least getting to their posts in the Legation. 

Notices were sent to all foreign residents in Peking by their respective 
Legations, instructing them where to assemble in the event of extreme dan- 
ger; green lights and guns were the designated signals. From the concen- 
tration points the foreigners would be escorted to the fortified Quarter by 
armed guards. I had arranged for my family to go to the American Lega- 
tion, but we had a carefully thought-out plan for the protection of my house 
and the great quantity of valuable equipment in the compound. With 
machine-guns posted on the roofs we should be able to present a pretty strong 
defense against any looters or even well-armed soldiery. 

The "City Fathers" did some sterling work in persuading the Fengtien 
generals not to let their troops come into the city, urging that foreign com- 
plications would arise with the inevitable looting. The soldiers encircled the 
walls, but with the gates shut and sandbagged not a man was allowed inside. 
In the meantime the social life of Peking proceeded much as usual. The 
only fish we could get came from the lake in the Forbidden City, fruit was 
non-existent, and there was a shortage of fresh meat. However, nothing is 
allowed to interfere with dinners and dances, polo and tennis. 

After a few days we discovered that Feng Yu-hsiang's army had retired 
up the railroad and entrenched themselves at Nank'ou, the pass through the 
Great Wall on the way to Kalgan. For many hundreds of years this pass 
acted as a strategic point protecting Peking from invasions of the Mongols 
and Tartars from the north. Now it acted in the opposite way by preventing 
the Fengtien army from following Feng's so-called "Christian soldiers" to 
their headquarters at Kalgan. 


The third evening of our siege I was amazed at the arrival of Mr. W. 
Douglas Burden of New York, a Trustee of the Museum, and Mrs. Burden. 
They had reached Tientsin a few days earlier, and I telegraphed them to 
wait there imtil it was possible to bring them up. They had discovered that 
two men in a motor car were attempting to get through to Peking and took 
the chance of coming with them. Fortunately for them, they picked up on 
the road the Fengtien general in command of the air forces, whose car had 
broken down. He was one of the few men for whom the city gates could be 
opened. I was at the Chi Hua Men gate talking to the soldiers when they 


asked me to please draw my car to one side as a very important general was 
arriving. The sandbags were being removed from one side of the gate, and 
when it swung open just enough to admit a car what was my amazement to 
see my friends come in ! They had had an adventurous trip up from Tientsin. 
The human heads hanging from posts along the road and the evil-looking 
soldier}' impressed them with the fact that their own lives were by no means 
safe. Their car was the first to come over the road, and Peking was keen to 
learn what was happening outside the city. 


I had had no news from Young since he reached Kalgan on the last train 
that ran from Peking. Telegrams would not be sent, and although wireless 
messages were accepted by the Chinese station in the Temple of Heaven 
they never were delivered. I tried every possible means of communication 
without success. Finally the American Consul in Kalgan got a radio through 
to the Consul General in Tientsin, who forwarded it to the Legation. It 
said that our caravan had finally started on April 26, after having been com- 
mandeered three times by the soldiers in defiance of the permit given by the 
General in command at Kalgan. He had lost so much "face" over the mat- 
ter that he furnished a military guard to see it beyond the limit of soldier 
activities. Young stated that he had tried to get down to Peking but had 
been turned back by the troops. 

The days dragged on interminably for all of us. Doctor Matthew had 
arrived and Granger and Nelson reached Peking from Szechwan after an 
arduous trip. All the staff were assembled with the exception of Young, 
who was in Kalgan. The cars were loaded in the courtyard of the head- 
quarters, and the Expedition was ready to leave at an hour's notice, but we 
were effectually blocked. Although the Fengtien troops were not pushing 
their advance against Nank'ou, heavy artillery fire could be heard every night 
in Peking. 

I tried to get permission to go by way of Shansi and reach the Mon- 
golian plateau west of Kalgan. Just when it seemed that it might be ar- 
ranged, Feng's army advanced along the proposed route. Therefore our 
plans came to an abrupt halt. There was a lull in the fighting at the Nank'ou 
pass and I entered into negotiations for our passage through the lines. But 
the Generals said that as there was a good deal of guerilla warfare going on 
they felt certain that we would be killed if an attempt were made. 


About two weeks later I was amazed to have a telephone message from 
Young. He had just arrived in Peking from Kalgan after a strenuous and 




adventurous trip. I went to the hotel and found him unshaven, gaunt and 
hollow-eyed. He had come by way of Shansi and had traveled about six 
himdred miles in order to get to Peking. Peking is only one hundred and 
twenty-fottr miles from Kalgan via Nank'ou. Young had walked a good 
part of a longer way, and the only food he could obtain were a few boiled 
and salted eggs. We got him into bed, and after a good many hours of sleep 
and some decent food he was fit and ready to do it over again if necessary. 

He reported that food in Kalgan was very low. For weeks the foreign 
residents had had no coffee, tea, sugar, milk or butter, and but very little 
flour. Occasionally they could get fresh mutton. Cigarettes were entirely 
gone. Fortunately the British-American Tobacco Company had a large 
garden in their compound and by forcing the vegetables under glass the for- 
eigners would be able to carry on for some time longer. Nevertheless, the 
situation was serious, for typhus had broken out, and the only doctor in 
town had died of the disease. 


Although Young had been able to get through both lines in Shansi, it 
was his opinion that it would be suicide to attempt it with six motors and a 
large party. We were certain, he said, to be annihilated by snipers even if we 
had permits and carried American flags. It wasn't good enough. Only one 
other possible route to the plateau remained; that was by way of Jehol, the 
old summer residence of the Manchu emperors, then over to Dolon Nor in 
Inner Mongolia, and down to Kalgan. Jehol is one hundred and forty miles 
from Peking, but the road is unspeakably bad. Lovell and I went in one of 
the Expedition cars, and I enjoyed the trip, for it took us through some of the 
most beautiful scenery in north China. A few miles beyond the An Tung 
Men, the north gate of Peking, we saw several thousand Fengtien cavalry 
coming across country at a sharp trot. They gave the appearance of retreat- 
ing soldiers, but we were not molested by them. 

Jehol itself is a beautiful spot and I was looking forward to enjoying the 
old palace and the tombs. But we found that the place was literally swarm- 
ing with soldiers who were so obnoxious that we remained there only a few 
hours. I had a letter of introduction from the American Minister to the 
Military Governor and, after a considerable wait, was given an interview. 
The Governor was a former brigand but received me courteously in a small 
room at the extreme right of the old palace. The walls above the kang, or 
bed platform, on which he sat were himg about with automatic pistols, and I 
noticed that one was always within reach. 

He would not even consider the suggestion of letting us through to Dolon 
Nor. He said that, were he to do so, he would be signing our death-warrant, 


for the region was so infested with brigands and deserting soldiers that it 
would require a very strong armed guard to take us through his own lines. 
Doubtless he was right, for in his outer office I had talked with two of his 
tax collectors who had been fired on by brigand soldiers while on their way 
to a station a few miles from Jehol. In the city we visited a British missionary 
and his wife, the only foreign residents. They said that for months they had 
been in a state of continual unrest due to the soldiers who swarmed through 
the place. Both of them seemed to be pretty well shaken nervously, and I 
can well imagine what they had been through. 


When Lovell and I returned to Peking, we had an unpleasant experience 
just outside the north gate. Although but few soldiers had been there when 
we went out two days earlier, the right wing of the Fengtien troops had re- 
treated from the Nank'ou Pass and were entrenching themselves across the 
road. In spite of our American flag we were stopped by soldiers who treated 
us as though we were brigands. Our passports were quite in order, but un- 
fortunately for us, the sergeant in charge could not read Chinese. For an 
hour this wretched guard kept us covered with loaded and cocked rifles while 
they searched the car and ourselves. In spite of my protests, the soldiers 
would neither take us to an officer who was able to read his own language, 
nor allow us to go. At the slightest motion on our part the boy soldiers (they 
were about sixteen years old) would throw their rifles to their shoulders. I 
indicated to our captor that the rifles might easily go off accidentally, in 
which case we woiold be killed. His only reply was "Mayo fadzu" (It can't 
be helped). 

At last an officer who had enough education to read Chinese appeared. 
As soon as he had examined our passports we were allowed to proceed. Had 
we been killed even accidentally, they would have reported that we had 
attacked them and they had been forced to shoot in self-defense. The soldiers 
showed quite plainly that they had spent a most pleasant hour in annoying 
two foreigners who, in front of cocked rifles, were helpless. 


There was always the hope that one of those sudden changes so frequent 
in Chinese politics would open the way for us, and it was not tmtil June that 
I had to finally admit that the 1926 Expedition must be abandoned. Doctor 
Matthew returned to New York via India, and Shackelford went across the 
Pacific. Captain Hill was assigned to the American Legation Guard in Peking 
to remain until conditions would allow us to proceed. Beckwith set to work 
studying Russian for the winter. 



Our caravan had gone as far as Shara Murun and there had awaited 
word from us. Hearing nothing, they returned to Hallong Osso, one hundred 
miles north of Kalgan. One of the Mongols, Tserin, came in to learn what' 
had become of us. He arrived just at the time when food in Kalgan was 
almost exhausted and the eighteen foreigners were facing real hardship. The 
American Consul managed to get a radio message to the Legation asking if 
they might take over our food. I immediately agreed and the supplies were 
brought to the city. The besieged foreigners sent us an enthusiastic note of 
thanks. It was not imtil mid-August that Feng Yu-hsiang's troops aban- 
doned the defense of Nank'ou and retreated into Shensi. The "Christian 
General" fled to Russia via Urga where he at once renounced all his Christian 
doctrines. Chang Tso-lin set himself up in Peking as Dictator of north China. 


We had long wished to make a reconnaissance of Yunnan from the stand- 
point of archaeology and palseontology. As conditions appeared to be nor- 
mally quiet there, Nelson and Granger prepared to spend the winter in that 
beautiful province. From the experience I had gained on my 191 6-1 7 Expe- 
dition, I was able to map out for them a tentative program, and they left in 
August for Ytmnanfu. 

Their previous winter's work in Szechwan had been interesting and val- 
uable. Granger obtained a splendid collection of Pliocene mammals from 
the fossil pits at Yen-ching-kou, which supplemented his former work. Nel- 
son was disappointed in finding no traces of palaeolithic man in the caves along 
the river. Most of them had rock bottoms. He came to the conclusion that 
the river had not been used as a highway of travel until man had advanced 
far enough to learn the use of boats. He did, however, obtain a considerable 
collection of neolithic implements and of those representing the interesting 
pre-Chinese culture first discovered by Dr. J. G. Andersson. 


As there was little that I could do in Peking during the winter, I sailed 
from Shanghai on September first for America. Our fruitless summer had 
cost considerably more than would have been expended dining a season's 
field work, and more money was urgently needed. I had also accepted an 
invitation to present the results of our explorations before the Royal Geo- 
graphical Society in London as the Second Asia Lecture on November 10. 
My winter was a busy one and added some fifty thousand dollars to the 
treasiuy of the Expedition. 



During the winter of 1926-27, important political events took place in 
China. The British concession in Hankow was forcibly taken by Chinese 
soldiers. To the stupefaction of all foreigners in China, the British Govern- 
ment not only made no attempt to reclaim it, but eventually officially returned 
it to China. Chiang Kai-shih, in command of the Southern army, was ener- 
getically making plans to attack the north with the object of bringing the 
entire country under the control of his party. He was ably assisted by the 
Russian, Borodin, and his army was directed by Russian generals. It had 
remarkable success and eventually moved on Shanghai with the openly avowed 
intention of taking that city and ousting the foreigners. 

Flushed by their entirely unexpected success in the Hankow affair, the 
Chinese had decided that the Foreign Powers would not protect their interests 
and that if force were used all foreigners could be driven out of the country. 
They demanded the return of the Tientsin concessions and of the Legation 
Quarter in Peking. Propagandists, directed by Borodin, were active in all 
parts of China and particularly in the Northern armies. The "battles," which 
Chiang Kai-shih was heralded all over the world as having won against the 
Northerners, were farcical. Propaganda had done its work so thoroughly that 
usually the enemy retreated upon the appearance of the Southerners, or else 
deserted to their side. 

The Foreign Powers had been driven too far by the advance upon Shang- 
hai, and war vessels and troops began to arrive from all quarters of the globe. 
The British sent battleships and a large force, the first upon the scene. It is 
generally admitted that their prompt action saved a most horrible wholesale 
massacre of foreigners. An indication of what would have happened aU over 
China was given at Nanking. A Southern army marched into the city, killed 
many foreigners, attacked the Consulates of several foreign powers, and began 
a systematic looting of every foreign house. Eventually the foreigners realized 
that they would be murdered to the last child and gathered on a hill belonging 









M O 


to the Standard Oil Company, where they were besieged by the infuriated 
soldiers. A gallant American Marine ascended to the roof of the building, and, 
while bullets were spattering all about him, signaled to the warships in the 
river. These immediately laid down a box barrage about the hill. As the 
first high explosive shell dropped near the house, the Chinese soldiers ran pell- 
mell in every direction. The foreigners were rescued by a landing party when 
the place had been cleared of Chinese soldiers. 

Meanwhile thousands of foreign troops had gathered in Shanghai. Barbed 
wire entanglements were erected about the foreign concessions and the city 
put vmder martial law. A curfew rule kept all residents indoors after ten 
o'clock in the evening. One or two clashes took place with considerable loss 
to the Chinese, but no determined attack was launched against the conces- 
sions. The Northern forces, having been completely disrupted by propaganda, 
retreated, and the Southerners occupied the native city of Shanghai. All 
foreign legations had ordered their nationals from the interior of China, for it 
was quite evident that the Chinese intended to repeat the Boxer attempt of 1900 
and kill or drive out every foreigner. Reports were continually arriving of mtir- 
ders and outrages committed upon foreigners in various parts of the country. 


Such, in brief, was the situation when I reached Peking in early April, 
1927. I came via Korea by train. Every ship out of China and every train 
was packed with departing residents, but coming to Peking I was the only 
foreigner on the entire train. At Tangku, not far from Tientsin, what was my 
surprise to see Granger and Nelson on the platform. They were just returning 
from Yunnan and we all came on to Peking together. Granger reported that 
their trip had been somewhat disappointing. They had first gone south and 
then east of Yimnanfu, because the western part of the province was so infested 
by brigands that the authorities would not allow them to enter it. They had 
discovered only one important fossil deposit; that was of Pleistocene age, but 
bandits drove them out after they had spent only three days there. Granger 
believed it to have most important possibilities and hoped to return for a 
further exploration. Nelson had fared little better. He had discovered no 
traces of Palaeolithic man but had iound further evidences of the Yangtze 
River Neolithic and pre-Chinese ciilture. I was sorry that they had not been 
able to explore the country to the west, for I believe that it will prove to be 
well worth investigating. 


Upon reaching Peking we foimd that the foreign residents were thoroughly 
frightened. It was the first time that I had seen anything like a panic. Even 


the year before, when the gates of Peking were closed and sandbagged, and 
Chang Tso-Hn's wild Manchu hordes were looting and burning the country- 
side, few foreigners in the capital were even nervous. The Hankow and Nan- 
king outrages, however, had inflamed the anti -foreign feeling which exists in the 
hearts of most Chinese. The poorly veiled hostility to foreigners by all classes 
made us realize that wholesale slaughter had been averted only by the arrival 
of foreign troops in China. The Southerners were pushing slowly northward 
and the Legations were already advising their nationals to leave Peking. 
Efforts were made to get as many of the women and children as possible to go 
to Dairen, Japan, or Manila. 


Suddenly the situation was completely changed by a dramatic raid by 
Chang Tso-lin's soldiers on the premises of the Russian Dal Bank, next to the 
Soviet Embassy. As Chinese soldiers are not allowed to enter the Legation 
Quarter, the raid was made with the permission of the Diplomatic Body. 
Although they had agreed to allow the soldiers to search only the Dal Bank, 
they went much further and ransacked the office of the Military Attache. 
The diplomats protested, of course, but only formally. The raid took place at 
eleven o'clock in the morning. I happened to be at the National City Bank 
on the opposite side of the street and witnessed the entire proceeding. It was 
most dramatic and totally tmexpected by the Russians. 

Until sLx o'clock that evening the searching proceeded. Many Chinese 
and Russian propagandists were unceremoniously hauled out from their 
hiding-places and hurried off to jail. One man attempted to bum various 
important documents, but the fire was extinguished before many had been con- 
sumed. Even in his wildest dreams, Chang Tso-lin could hardly have believed 
that his raid would produce such important results. Evidently the propa- 
gandists had depended upon the diplomatic immunity of the Embassy and 
had used it as a central clearing-house from which operations were conducted 
in various parts of the world. Most incriminating documents were found. 
Few people realize that raids which subsequently took place in London, Paris 
and in Argentina were made upon information obtained at the Soviet Embassy 
in Peking! 

Chang Tso-lin then set to work systematically to rid north China of the 
propagandists. Those Chinese who were caught in the Embassy raid were 
slowly strangled. Search parties were busy day and night rounding up the 
others whose identity had been disclosed by the capttu-ed papers. Hardly a 
day passed that one or more persons were not executed at the public ground 
opposite the Temple of Heaven. The place has the appropriate name of "The 
Heaven's Bridge." The propagandists fled from north China like rats deserting 


a sinking ship. In a very few days Peking and its environs had lost its appre- 
hension and settled down into qmte a normal existence. Heavy fighting was 
going on, but it was still some distance from Peking, and the raid had so dis- 
rupted the system of the Southerners that Chang Tso-lin made some progress 
in driving them back. Finally, however, they recovered from the blow and 
again began to advance slowly northward. 


The prospect for continuing our explorations in Mongolia could not have 
been blacker. Even had I been able to get the Expedition away from Peking, 
the American Minister would have prohibited us from leaving. Hardly a 
foreigner was left in China away from the seaports. Still, it was most dis- 
heartening to face another season of inactivity. 

Granger, Olsen and Nelson sailed for America. Only McKenzie Young 
remained with me at the headquarters. We proceeded to liquidate certain 
effects of the Expedition, put others in a place of comparative safety, and re- 
duce current expenses to the minimum. I decided to stay in Peking during 
the winter, hoping that one of those sudden changes that so frequently occur 
in Chinese politics would give some encouragement for an expedition in 1928. 


When I arrived in Peking, I found Dr. Sven Hedin, the famous Swedish 
explorer. He had come during the winter with a large staff of Swedish and 
German scientists and comprehensive plans for an expedition by air across 
central Asia. His work was largely financed by the Luft Hansa, a German 
concern, and was to include other sciences besides those relating to air navi- 
gation. He had met most tinexpected opposition in Peking. An anti-foreign 
group which called themselves "The Society for the Preservation of Ciiltural 
Objects" had launched a bitter newspaper attack not only upon Doctor 
Hedin but upon all foreign expeditions. They claimed that China was being 
"robbed of priceless treasures" even when such "robbery" was limited to 
meteorological observations, geologic specimens and fossils. The Society was 
an entirely unofficial organization, but by false publicity it had succeeded in 
arousing such popular indignation that the government authorities dared not 
ignore its activities. After months of delay Doctor Hedin was forced to 
abandon his program of aeroplane exploration. He endeavored to save some- 
thing out of the wreck by proceeding by camel and carrying on other forms of 
scientific investigation. Even then the Cultural Society insisted that a 
Chinese be appointed as co-director of the expedition and that all of the 
collections be given to the Chinese. In addition he was forced to accept ten 
Chinese students and "professors" as members of his party. 


Rather than abandon his entire expedition, Doctor Hedin at last agreed 
to these extraordinary demands. While I thoroughly understood and sympa- 
thized with Doctor Hedin's position, it was obvious that his acceptance of such 
preposterous conditions would make it extremely difficult for other expeditions 
to carry on their work. My predictions were amply justified by future events. 
By the time Doctor Hedin was allowed to leave, conditions in the interior had 
improved sufficiently to make it safe to go with a large party such as his. It 
was then much too late for our expedition to start, since our working period 
was of necessity limited to the siammer months. 


I spent the entire winter in Peking, watching events most closely. The 
war had developed into a stalemate and neither side was making important 
gains. It became quite evident that the real test of strength would come in the 
next svunmer, 1928. By the end of January I had made up my mind that we 
would make a final effort to get into Mongolia. I hoped that we should be 
able to slip out during the interval of comparative quiet between the end of the 
winter and the beginning of the summer campaigns. If we were caught again 
I determined to abandon the work. After cabling the Museum to have all the 
staff reach Peking not later than April 14, I set about collecting a new caravan, 
for I had sold our camels to Doctor Hedin. It was extremely difficult to buy 
camels because so many had been confiscated by the soldiers. Reverend Joel 
Eriksson, of the Swedish Mission at Hatt-in-Sumu, one hundred and thirty- 
five miles north of Kalgan, consented to act as my agent. Slowly he collected 
the one hundred and twenty-five camels necessary for oiu" work. 


Conditions on the Mongolian plateau at that time were extremely bad. 
Because of the imceasing warfare, virtually all business with Mongolia had 
been suspended for nearly two years. The entire region swarmed with bandits 
who raided right up to the walls of Kalgan. If an automobile or a caravan 
left the city it was robbed before it had gone fifty miles. The brigands held 
complete sway over all the cultivated region through which it was necessary to 
pass before one could reach the desert. 

Eventually conditions became so bad that something had to be done. 
Not only were the merchants being ruined, but the bandits themselves were 
dying of starvation. There was no one to rob and the food in the border villages 
had all been consumed. As usual in China, the Chamber of Commerce in 
Kalgan took matters into their own hands and entered into negotiations with 
the brigands. The local government officials agreed that certain "liaison 
bandits" would be allowed to enter Kalgan and make their own private 


arrangements with the caravan owners. Those caravans that paid five 
dollars a camel wotild be allowed to pass through the brigand area 
unmolested. Each motor car was to pay one hundred dollars. This form 
of "protection" is common in China and works smoothly enough as a rule. 
Early in February, after the arrangements had been completed, thirteen 
thousand camels left Kalgan at one time bound for Urga, Uliassutai, Kobdo 
and Hami. 


Meanwhile McKenzie Young and I collected supplies and equipment in 
Peking and transported them to Kalgan. We were forced to pay exorbitant 
transit duties on everything. Taxes were levied when oiu* stuff left Peking; 
again when it was half-way to Kalgan; again when entering and still again 
before leaving Kalgan. Such taxes amounted to nearly three hiondred dollars 
on each motor car alone. We had eight cars. All of them had been in and 
out of Peking and Kalgan several times and should have been exempt ; we had 
the former tax receipts, but that made not the slightest difference. If the taxes 
were not paid in full, the cars could not move. The new officials who had 
obtained posts did not expect to hold them long, since everyone felt that a 
change in the political situation would take place during the summer. There- 
fore they were all trying to make as much money as possible out of 
their positions while they could. In addition to the so-called official 
taxes, nearly all the offices swarmed with individuals who would detain our 
equipment or make other trouble luiless they were given a generous amount 
of "squeeze." It cost us about ten thousand dollars in taxes and squeeze 
to get our cars, camels, food and equipment out of Peking, and into and out 
of Kalgan. 


Before we started to ship our equipment, the American Minister, Honor- 
able J. V. A. MacMurray, had called upon the Dictator, Marshal Chang Tso- 
lin, and informed him of otir desire to continue work in Mongolia. The Marshal 
had given his consent and I had gone to Kalgan to make arrangements with 
the local authorities as in the past. I found them agreeable and they gave the 
necessary permits for us to leave for Mongolia. I knew, however, that it 
was imperative to have no publicity about our departure, because it might 
stir into action the unofficial Cultural Society which had caused Hedin so much 
difficulty the previous year. To this end the foreign correspondents of various 
world newspapers and those of the local press acted with the greatest kindness. 
They all agreed to make no mention of the Expedition until after our departure, 
and kept their promises to the letter. 



Our caravan reached Kalgan early in March, with five thousand other 
camels just down from Urga. Accompanying them was a bandit "liaison 
officer." A few years before he had been a respectable landlord of one of the 
motor inns on the Urga trail. I knew him well and knew that now he was a 
head brigand. What is more, he knew that I knew it. But it would have 
made him "lose face" to admit the fact. Therefore, we were introduced as 
though we had never seen each other. While he remained in Kalgan he posed 
as a "general" who could arrange protection for our caravan through his 
"soldiers." Half an hour of tea-drinking and extraneous conversation ensued 
before we got around to business. He suggested the customary fee of five 
dollars a camel. I offered one dollar. He knew that ow boxes contained noth- 
ing that his brigands could use or sell, and eventually we settled on half the 
usual rate. 

Just before he left he brought up the matter of the one-hundred-dollar fee 
for each of our motor cars which would go out some weeks after the camels had 
left. I became very vague at that. We were vmcertain when we would leave — 
how many cars there would be — I would get in touch with him later! But 
I did mention that after all we hardly needed protection for we would have 
thirty men on the cars, all would have rifles and, moreover, there would be a 
machine-gun which could shoot two hundred bullets a minute. I did