Chapter I: The Ordos Border

Very early in life my travels began. I was only four years old when my parents took me from an interior town of China to the coast, and thence to Europe. Vaguely I remember the mule train “coughing in the dust” of a Shansi road, and the house-boat journey down a Chihli river, with my mother lying at death’s door in the cramped and tiny cabin, overcome by that cruel journey. Then came the strange sights and scenes of the sea journey to England, followed by the even stranger (to me) life of my native land. Three years later we were back again in the interior town, and from that time on I have scarcely lived in one place longer than a year or two at a time.

All this doubtless accounts for my love of travel, which has led to my kicking the traces and becoming a wanderer. What more natural than that the wandering should have for its scene the land of my nativity, where I know the people and their language, where I can live, if necessary, upon the land, without the help of stores from Europe or American canned goods? What more natural than that the unknown places in this land of mystery should beckon to me, and that I should hear that voice ringing its “interminable changes ... something hidden. Go and find it....”?

I cannot claim with Kipling’s Explorer to have found country unknown to man; but I can say that I have been the first white man to tread many a forest and wild mountain, many a sandy waste and boulder-strewn wilderness, many a rich pasture and fertile valley. I, too, have seen the promise of future prosperity in “the big fat marshes that the virgin ore-bed stains,” in the “nameless timber,” and “illimitable plains.”

Will others go up and occupy? Will my countrymen aid in developing that potential wealth? I hope so.

Strange though it may seem it has not been the solemn grandeur of the great back ranges, nor the mysterious silence and gloom of the virgin forests, nor yet the smiling fatness of the valleys and plains that have appealed to me most. It is the sun-baked, barren ridges, the shifting, windswept sand-dunes and the saline, brackish swamps of the Ordos Desert that have cast upon me the strongest spell.

Even as a boy I had been fascinated by what I had heard of that howling wilderness, that waste of sand, that “Never-never country.” The romance of those wild Tartars, living their lives of untrammelled freedom always appealed to my imagination and I longed to go and see it all.

At last the chance came, and never shall I regret the step that once and for all settled my destiny, when I exchanged a town life of comparative ease for that of the explorer and collector.

It was in Tientsin in the summer of 1907 that I first met Mr. Malcolm P. Anderson, who was out in the East collecting mammals and specimens of zoological interest generally for the British Museum at South Kensington, in connexion with the Duke of Bedford’s Exploration of Eastern Asia.

At the time I was engaged in mounting the fruits of a hunting and collecting trip in Western Shansi for a local museum.

With common interests a friendship sprang up between us, which culminated a month or so later in a decision to join forces in carrying out an expedition into the hitherto practically unknown Ordos Desert.

As I would not be free till the end of the year Anderson went to T’ai-yüan Fu in Shansi, where he continued to collect, till, everything having been arranged satisfactorily, I joined him towards the end of January 1908.

The proximity of the Chinese New Year prevented our starting for another fortnight, but at last on February 10, we set out from T’ai-yüan Fu with a somewhat meagre equipment of four mules and one pony, together with three natives.

A glance at the map of China will suffice to show the reader the position of the Ordos Desert. It is situated within a mighty loop of the Yellow River immediately north of Shensi, from which province it is divided by the Great Wall. The Ordos is known to the Chinese as Ts’ao-ti (grass land), and the only information we could glean concerning it was that it was an immense wilderness of sand and coarse scrub, in which wild Tartars lived, and where hardy races of small ponies and cattle were bred.

This sounded interesting, though hardly favourable to our purpose of making a large collection. Nevertheless our instructions were definite on the subject, so we set about attaining the end in view as best we could.

We found that by travelling due west we could have reached Yü-lin Fu on the border of the desert in fourteen days, but this would have brought us to the country at a time of year when the weather would have been most unsuitable for the collecting of small mammals, and much valuable time would have been wasted.

PLATE I.

[_Photo by M. P. Anderson._

YEN-AN FU LOOKING UP THE YEN-SHUI.

_Opposite p. 6._]
]

We therefore decided to travel south-west to Yen-an Fu in North Central Shensi, and from there work north-west, to the southernmost point of the Ordos. By stopping and collecting at a few places along this route, we would gauge our arrival on the frontiers of the desert to sometime in spring, the best season for our work.

After leaving T’ai-yüan Fu, and crossing the low mud-flats of the Fên River, our road lay in a south-westerly direction through well cultivated country. Round the villages we noticed innumerable jujube trees (_Zizyphus sativa_), which grow extensively in all the low lying valleys and plains of Shansi, Chihli and Shensi. The fruit of these trees resembles nothing so much as a date, and when treated with honey makes an excellent preserve. It is largely used in the manufacture of sweetmeats, especially of the cheaper qualities, while bread-steamers[1] use it instead of currants in the manufacture of a kind of bun loaf.

Our first stopping place was the small town of Chin-ssŭ, famous for its temple and spring. The latter gushes out of the base of the hill upon which the temple is built, and irrigates the surrounding fields, making the cultivation of rice on a large scale possible.

It also forms the motive power of a number of paper mills, a coarse straw-paper being manufactured in the district.

The temple of Chin-ssŭ is a very beautiful edifice, composed as it is of magnificent buildings standing upon a series of stone terraces running up the steep hillside. A long flight of stairs extends from the base to the topmost shrine, while in the extensive grounds are artificial lakes, filled with carp. Old and stately trees stand here and there, the most interesting of which are two huge Maidenhair trees (_Gingko biloba_).

We reached Fên-chow Fu two days later, passing on our way the important towns of Chiao-ch’êng Hsien and Wên-sui Hsien. The country between these two places is noted as being the finest grape producing district in China. The grapes grown are mostly purple, but some white varieties occur. In the season these fruits sell at less than one halfpenny per pound.

Fên-chou Fu itself is a large business town with an extensive floating population. It has nothing of particular interest to recommend it to the traveller, who is only too anxious to get away from its dirt and smells.

On leaving this town we took a westerly course, and following a deep and narrow gorge in the limestone formation, crossed a high divide at its head.

We spent the night at a large village named Wu-ch’êng. From here we turned southward, and after travelling fifteen miles, stopped at a place named Shi-shi-ling, situated in a mountainous and well-wooded district. Here we remained for some days hunting and collecting.

Wild boars, leopards and roe-deer were seen, but we were too busy trapping for small mammals to pay much attention to large game.

Pheasants, partridges and hares were abundant, and kept our pot well filled. The weather was too severe for good work, and beyond a few squirrels, field-mice, small hamster rats and three or four species of birds we secured very little.

The country was typical of Shansi, consisting of rugged shale and limestone mountains, overgrown with scrub oak, hazel, birch, wild peach, conifers and thorny shrubs of various kinds.

From Shi-shi-ling, we travelled westward towards the Yellow River, putting up at Ning-hsiang Hsien, and Liu-lin-chên on February 24 and 25 respectively. The former of these two places is a town of no importance and less interest. At Liu-lin-chên some fine springs, open all the year round, form a winter resort for countless wildfowl, including teal, mallards, sheld-ducks, golden eyes and mergansers.

On February 26 we crossed the Yellow River and put up at a small village on its right bank. The crossing was effected by means of large ferry boats, into which mules and baggage were ignominiously bundled, when, with long, roughly made oars, the ferrymen kept the boats at angles such that the swift current forced them along in the desired direction. In consequence, the landing was effected some considerable distance down stream.

Continuing westward over barren windswept loess ridges for two days, we reached a busy little town named Sui-tê Chou. Thence we travelled southward for four days along a splendid road, which had recently been cut by the Yü-lin Fu garrison. This country was very desolate, consisting as it did at that season of bare loess hills inhabited only by occasional coveys of partridges and a few foxes and eagles.

At a place called Yen-ch’uan Hsien, not far from some recently opened oil wells, we turned westward, and after crossing a steep loess pass, entered the valley of the Yen-shui, the river on whose banks is situated the old town of Yen-an Fu. All the towns and villages we passed were in a sad state of ruin, while the inhabitants were terribly poverty-stricken. The country has not yet recovered from the devastating effects of the great Mohammedan rebellion of the sixties, followed in 1887 by a protracted famine. It is true that the greater part of the loess hills is under cultivation, but years must elapse before the country regains its former prosperity.[2]

Near Yen-an Fu we noticed that there were many dense thickets, consisting chiefly of the thorny wild jujube (_Zizyplus sativa spinosa_) along the Yen-shui valley, so we decided to stop and spend a few days in collecting. Our efforts met with immediate success, so that we prolonged our stay in the vicinity. We secured one new species and three new subspecies of rodents, besides several already well-known forms.

Each morning ere it was light we would sally forth with satchel and shot-gun to inspect our traps, set the night before. Bringing in the specimens secured, we would spend the rest of the day till mid-afternoon in skinning and preparing our day’s haul, after which we would once more go out to look over the traps, or set new ones.

The new species referred to above was a pika (_Ochoton abedfordi_), a small rabbit-like rodent, while the new subspecies were a sulphur-bellied rat (_Epymys confucianus luticolor_), a large hamster rat (_Cricetulus triton incanus_) and a hare (_Lepus swinhoei subluteus_). The kindred forms of the last three species are found in the Shantung Peninsula and in Chihli.

The abundance of game and small mammals round Yen-an-Fu is undoubtedly due to the extensive thickets of wild jujube in the vicinity. This plant, wherever it occurs, forms excellent cover for pheasants, hares, small mammals and birds. It grows usually in the form of low shrubs, and is thickly covered with strong sharp thorns, some of which are straight, and others hooked. The result is that large animals and human beings experience great difficulty and inconvenience in traversing thickets of this growth. The fruit, which has a pleasant sour taste, together with the hard stony kernels, forms a regular part of the autumn and winter diet of the smaller mammals and birds. One of the characteristics of this peculiar plant is that it never grows at an altitude exceeding 3,000 ft. above sea level, and another is that it comes into leaf long after all other plants in the same districts in which it grows.

We were told by the natives that there were some very good collecting grounds about ten miles south-west of Yen-an Fu. Accordingly we packed our traps and visited the spot. We were rewarded by finding excellent country, both for shooting and trapping, so that our collection began to reach pleasing proportions.

This country we found to consist of the usual loess hills, but they were thickly covered with scrub and rapidly extending woods of birch, hazel and poplar.

In some places the hillsides were covered with wild peach (_Amygdalus davidiana_), which was now in full bloom, and presented a wonderful sight. This peach is supposed to be the original stock from which all our many varieties have been derived. This however is not at all certain, as wild peaches have been found in other parts of Asia. It is very hardy, withstanding extremely low temperatures. The fruit is small and worthless as a table delicacy, though it is greatly appreciated by squirrels and other rodents.

The tracks of wild boars, roe-deer, leopards and wolves were frequently seen, while pheasants were particularly plentiful.

Some interesting birds were secured, amongst which were a large and handsome blue magpie (_Urocissa sinensis_), and the sweet-voiced timeline (_Pomatorhinus gravivox_). A single specimen of the grey-headed plover (_Microsarcops cinereus_), a large and elegant member of the family, was shot in one of the stream beds.

The weather was now wonderfully mild, so that our thoughts began to turn towards the sandy stretches of the Ordos, and finally on April 6, after once more reaching Yen-an Fu, we set out in a north-westerly direction along the Yen-shui valley.

We were somewhat reassured as to the prospects of securing food for ourselves and our animals in the Ordos, the natives of Yen-an Fu having informed us that there were plenty of Mongols, who were well off and hospitable, scattered in little encampments at easy stages in every direction.

The journey up the Yen-shui proved uneventful except that as we neared the end of the valley on the third day, we had to cross the river no less than sixty-eight times. Towards the end of the afternoon, however, it began to dwindle down till it was no longer of formidable dimensions and crossings were easily effected.

As the day wore on we noticed that the loess hills were growing higher and higher, while the substratum of rock changed from grey to red sandstone. Through this the stream cut deeply, forming sheer and lofty cliffs on either side.

In one place we noticed, high up on the face of the cliff, some cave dwellings, which were reached by a series of stakes driven into the face of the rock. Many of these were missing, from which we concluded that the caves had long remained untenanted.

PLATE II.

[_Photo by M. P. Anderson._

CAVE DWELLINGS NEAR YEN-AN FU.

_Opposite p. 14_.]
]

The natives of our last stopping place told us that here we would find a good-sized village with plenty of accommodation. This turned out to be a purely malicious and unprovoked falsehood, for when night began to descend our little party found itself in a desolate and seemingly uninhabited wilderness.

Following the only bridle path we began a long and difficult ascent till we reached an altitude of 6,000 feet, having climbed over 3,000 feet since leaving Yen-an Fu. The path now led along the top of a rugged ridge and became momentarily more difficult, till, darkness overtaking us, we realized that we were hopelessly lost. To make matters worse, a cold north wind sprang up, cutting us to the bone.

We wandered aimlessly along for two hours, discussing the advantages or otherwise of making a tentless camp where we were, without fuel for a fire, food for ourselves, fodder for the mules, or even a drop of water.

Just as we were about to unload the mules a voice hailed us from the darkness, and the welcome barking of dogs told us that we were near some human habitation. To our joy we found not fifty yards from us a cluster of huts, including an inn, surrounded by a high mud wall. The latter had effectually hidden the lights of the huts, and but for the vigilance of the innkeeper we should have been condemned to spend the night exposed to hunger, thirst and a bitter north wind.

The following day we pursued our course over steep mountain peaks, down frightful chasms and treacherous cliffs, and over wastes of sinking sand, but at last we reached Ching-pien Hsien, a small garrison town within a mile of the Great Wall, and the longed for Ordos Desert.

However pleased we might be at having reached our goal we were nevertheless far from comfortable. The weather had turned very cold again; nothing but the very poorest food was obtainable. The air was filled with fine sand, which got into our eyes, scrunched between our teeth, and stung our faces as it was driven by the cruel north wind.

“... I think never saw
Such starved ignoble nature; nothing throve:
For flowers—as well expect a cedar grove!
But cockle, spurge, according to their law
Might propagate their kind, with none to awe,
You’d think; a burr had been a treasure-trove.

· · · · ·

As for the grass, it grew as scant as hair
In leprosy; thin dry blades pricked the mud
Which underneath looked kneaded up with blood.”

The utter dreariness and desolation of the surrounding country, together with the poverty and wretchedness of the inhabitants, combined to depress and harass our spirits, which were only restored to their normal condition by the assurance that the country beyond the Wall was much better. We made a short excursion as far as the Wall, and climbing one of the isolated towers, obtained our first view of the Ordos.

A wide expanse of sand, thickly studded with trees gave promise of better things in store.

On the following day, April 11, we lost no time in leaving behind us the miserable town of Ching-pien, and for some miles travelled northward through a country that grew more promising with each succeeding step. Flocks of stately cranes fed on the stretches of reddish sand, while innumerable trees, chiefly willows, arranged in long rows, relieved our eyes, after the days of barren wastes we had endured.

In crossing a deep ravine, just after leaving Ching-pien, we nearly lost one of our mules with its precious burden, comprising our only stores, in some deadly quicksands. It was only a miracle that saved the animal. As it was much of our food got wet, and some of it was rendered useless.

That evening we reached the end of the broad belt of trees, and put up at a small farmstead named Ning-t’ieh-liao, which we were surprised to find was owned by Chinese.

We learnt that the Chinese here, and elsewhere along the Ordos border, are gradually pushing further and further into the Mongol territory.

By planting rows of trees they prevent the drifting and piling up into dunes of the loose shifting sand. They manage thus to grow crops of millet, such areas as they do not cultivate producing a coarse grass. This with the leaves of the willows forms the chief food for their flocks of sheep and goats.

We spent April 12 at this farmstead, drying out our stores and exploring the immediate vicinity. We found that the river that flows past Ching-pien continues northward into the desert, subsequently turning eastward and flowing back into Chinese territory.

Being anxious to see something of the Mongols themselves we left Ning-t’ieh-liao the following day, taking a northerly direction.

We were told that the Tartars lived some twelve miles away, and that it would be necessary to hire a guide to lead us across a difficult stretch of sandhills. We therefore engaged the services of a stupid yokel, who had accompanied us from Ching-pien, and who claimed to know some of the Mongol families.

We soon left behind us the belt of willows, and passing a small lagoon, from which Anderson secured a fine grey-lag goose, we entered the dunes. Considerable difficulty was experienced with the mules, whose small hoofs sank deep into the sand. Spills were frequent. Once all four mules were down together, piling one atop of the other as they rolled down a more than usually steep slope. Great confusion followed, and it was only with the greatest care that animals and loads were extricated without serious damage.

We also found going very difficult and tiresome, owing to the looseness of the sand.

At last after we had passed through a belt of tall scrub, there opened out before us a long and broad stretch of grass land dotted over with lagoons and marshes. As we approached these we noted vast numbers of waterfowl of every description. Grey-lag geese and sheld-ducks wandered in pairs between the tussocks of coarse grass in search of food. Flocks of stately cranes and lordly bustards waded knee deep in the marshes, or fed upon the grassy meadows. Teal and ducks of every variety blackened the surface of the lagoons, while plovers flapped heavily overhead uttering their plaintive calls.

Scattered far and wide over the plain were innumerable herds of ponies and cattle, the property of the Mongols whose huts and tents we could see on the opposite edge of the grassland.

We made our way across to one of these huts, where we interrogated the goodwife, who was brought to the door by the barking of some savage and wolfish looking dogs. She refused us shelter on the ground that she could not supply our mules with the necessary fodder.

We passed on to the next hut and met with a similar reception. At last after trying several farmsteads we reached a newly built, two-roomed hut, enclosed in a neat willow fence, and, unloading our mules we refused to proceed any further.

Finding that these border Mongols could speak Chinese fluently, I went into the hut to discuss with the owners of the place, a newly married couple, the much vexed question of quarters for ourselves and fodder for our mules.

The Mongols were polite, and offered us shelter for one night only, this being the custom. More they dared not do, or they would be held responsible for our actions. It was plain that they were suspicious of us, and my explanation that we were collecting small animals only seemed to deepen their suspicions. Evidently they had not yet experienced the eccentricities of white men.

At last we decided to take what we could get, and as we could not bring ourselves to sleep in the hut with the natives, we put up our beds inside the willow fence.

We were next invited to take some salted tea, in which parched millet was steeped. This was not very palatable, but was better than nothing.

The hut was built after the manner of those occupied by the poorer Chinese, and, as already stated, was enclosed by a high willow fence. There were also stables and storehouses built of willow stems plastered over with mud. The confines of our hosts’ property were marked by a low mud wall, about two feet high, outside which was a ditch of corresponding depth.

A well, some six feet in depth, supplied the residents and their cattle with water, that of the lagoons in the district being too brackish for either man or beast. Even the wells were far from sweet.

Outside and opposite the gate of the enclosure was a long pole stuck into an immense wicker basket filled with sand. From a crosspiece near the top of the pole depended several pieces of white and red calico, while the end of the pole was adorned with a tuft of dry grass. The whole contrivance formed a sort of totem pole, having something to do with the religious ceremonies of the owners. We noticed one of these poles outside every Mongol hut or tent.

The tents of these Mongols were of the usual pattern—wooden framework covered with coarse felt, usually of a black colour.

We soon found that these people were intensely superstitious, practising a sort of demon worship. Everywhere, they said, were malign spirits, which must not be offended. Thus I was called sharply to book, when I inadvertently sat upon the coping of the mud fireplace. It was insulting the “fire demon.” Again, as we approached one of the huts, the muleteers led their animals across the low boundary wall, when an old woman rushed out and cursed us roundly, while her daughter tore the bells off the mules. Completely discomfited by the vigorous onslaught we retired in disorder, afterwards learning that some religious function in connexion with the birth of a child was in progress, and the ringing of the bells within the confines of the farmstead would attract evil spirits.

At one place we were refused admission on the ground that they were having a spring cleaning of devils, and if we entered we would assuredly bring some with us. Even at the place where we did secure quarters our baggage was left outside the enclosure long enough for the demons it contained to escape.

Our hostess on finding a snake inside the mud wall very carefully carried it outside, and refused to let us have it for a specimen, saying that it would not do to offend the snake spirit.

For the rest these people are Buddhists, one or more of the sons in each family becoming lamas.

In physique the Ordos Mongols are superior to the Chinese. They are a tall well built race, admirably fitted to wrest a living from the cruel desert in which they live. The women are good looking, some even beautiful, while their lives in the open give them a grace of carriage together with a strength and freedom of movement never seen inside the Wall. They can ride barebacked, and with the ease and grace of a cowboy, the shaggy and wicked little ponies that they rear upon the surrounding grasslands.

They wear a peculiar headdress of silver-mounted coral and turquoise beads, surmounted by a gold embroidered cap, while their hair falls over their shoulders in two thick strands. The rest of their ordinary attire is similar to that of the men, which is the usual Manchu dress—baggy trousers and long coat or gown of blue surmounted by a large sleeveless waistcoat of some bright colour. Only in the absence of buskins does the women’s dress differ from that of the men.

On gala days, which the women celebrate equally freely with the men, the former sport silk robes of the most gorgeous and vivid colours, and even the latter manage to don clothes of gayer hues.

They are a light-hearted, cheery people, always on the look-out for fun, and always playing jokes upon one another.

As already stated a large proportion of the men are lamas, who have their heads shaved and wear long robes of red or yellow, with sashes of the same colours. These priests are supposed to be celibate, but we saw several with large and flourishing families.

The food of these Mongols consists in winter of mutton and beef, but in summer they live almost entirely upon parched millet, which to render palatable they steep in salted tea. The millet, a poor and coarse variety, they grow themselves.

For fuel they depend almost entirely upon the droppings of their cattle, which when dried they call “argol.” Camel “argols” are undoubtedly the best fuel of this nature.

While setting some traps in the willow scrub that fringed the grasslands, dividing them from the wastes of sand beyond, we noticed great numbers of hares, which, as dusk came on, left the sheltering bushes, and frolicked or fed upon the meadow land. We also saw a few pheasants and partridges.

Next day, having brought in our traps and packed our baggage we loaded the mules, and were about to start, when up rode a well dressed old man, wearing a mandarin’s hat and blue button, and accompanied by two attendants. Jumping from his pony he saluted us, producing a crucifix to show that he belonged to the true faith. After exchanging civilities we told him of our difficulty in finding quarters. He at once gave us permission to stay where we pleased as long as we chose. Our erstwhile hosts promptly offered us further hospitality, but with such a charter as we now had we preferred to go further in search of better collecting grounds and more commodious quarters. The old chief offered us rooms in his own home, which he said was at a place named Borrobalgosun some twenty miles westward. Accordingly we set out together in that direction and soon were engaged in friendly conversation.

We found that our new friend thought that we were Catholic priests, of which religion he was a convert. On finding that we were not even missionaries of any other sect, he changed his attitude towards us. He was still polite, but less _bonhomme_ and presently he withdrew his invitation to stay at his home.

However he promised to find us quarters elsewhere, and excusing himself on the plea that he had some business to attend to, set his pony at a gallop and was soon out of sight, leaving us to follow as best we could.

At last after passing several more farmsteads, we came to a cluster of huts, outside which our friend awaited us. We were well received, being at once shown into a spare room, and offered salted tea and millet.

Mr. Bluebutton introduced us to our host, and all arrangements were made for a stay of some days, after which, said he, we would be welcome to Borrobalgosun. Things having thus been arranged to his entire satisfaction, the chief bade us a hearty farewell, and with his minions rode away into the west.

We at once set about making ourselves as comfortable as possible. We got out our traps, but met with little success. The only things caught were specimens of ground squirrels (_Citellus mongolicus_) and two varieties of gerbils (_Meriones unguiculatus_ and _M. auceps_), together with the hare.

An attempt to secure some antelopes failed, though quite a number were seen. Specimens of a snake (_Coluber dione_) and two lizards (_Phrynocephalus frontalis_ and _Eremias argus_) were added to the collection. Besides the waterfowl and game birds already mentioned, wheatears and larks seemed to be the only feathered denizens of these sandy wastes.

The scarcity of animal life was a great disappointment to us, and we decided to waste no more time in this locality. The muleteers came to us complaining that they could get no grain for their mules, while very little fodder was to be had. The obstinate animals refused the coarse alkali-impregnated herbage that grew around. Our own stores also were sadly diminished, and we were told that we would not fare better by going to Borrobalgosun, where nothing but parched millet was to be had.

Under these circumstances we decided to head for Yü-lin Fu, where we could replenish our stock of provisions, and properly equipped could once more attempt to penetrate the desert.

Five days we travelled in a north-easterly direction across heartbreaking sand dunes. By the end of the first day we had left behind us the Tartar encampments, and found ourselves once more in country occupied by Chinese.

The villages we passed were few and far between and we suffered considerably from thirst and heat. The white sand reflected the blinding rays of the sun, which shone ever more fiercely, till our eyes ached and our faces scorched and peeled.

From time to time we crossed shallow rivers, all of which flowed in a south-easterly direction. They had cut down through the fifty feet or so of sand to bedrock, which consisted of the same red sandstone that we had noticed at the sources of the Yen-shui. At one place there was a wonderful natural arch spanning the river, over which the road lay.

On the third day, April 18, we reached a place named Hai-liu-t’u-ho, where we secured the first specimen of the Ordos jerboa (_Dipus sowerbyi_).

Herbage of any sort was very scarce, dwarf willows, a coarse grass and a sort of sage brush composing the sparsely distributed thickets.

In one or two places we came across a species of conifer, which grew in thick clusters amongst the sand hills, and which we found to be very resinous, burning furiously even when quite fresh and green.

April 19 we spent at Hai-liu-t’u-ho in the hope of securing some more jerboas. Two more were brought in by the natives. That day a terrible sandstorm sprang up, so that when we set out the following day there was no sign of a road. We could not hire a guide as the people refused to have anything further to do with us. A baby and a cow had died in the village during our stay, and the superstitious natives looked upon us as the cause of these disasters.

The consequence was that we soon went astray. Another sandstorm commenced, and before long we were hopelessly lost. For hours we wandered over the dunes with blinded eyes and sand-choked lungs, but at last we found a desolate looking farm, the owner of which we induced by promise of reward to set us on the right road once more. This he did, and we were able to keep to it by following some newly made tracks.

The following day we reached Yü-lin Fu, a large border town, situated on the bank of the Yü-lin Ho, some three miles south of the Great Wall. Here we found comfortable quarters, and thankful we were to escape for a time the all-pervading sand.

We at once set about making plans for another trip into the Ordos, but first we decided to offer money for specimens. A notice to the effect that the foreigners were willing to buy wild animals was put up at the South Gate, where all the country people and market gardeners would see it.

Within a few days we reaped a rich reward, so much so that we abandoned any idea of making another journey into the desert. Jerboas, hedgehogs, hamster rats, molerats, chipmunks and ground squirrels were brought in in great numbers. Of these the jerboa (_Dipus sowerbyi_), the hedgehog (_Erinaceus miodon_) and the sand hamster (_Phodopus bedfordiæ_) proved to be new species, while the chipmunk (_Eutamias asiaticus ordinalis_) was a new subspecies. Besides these a fine badger (_Meles leptorhynchus_) and a wild cat’s skin were purchased.

One day we visited a peculiar fort guarding a gate in the Great Wall, through which once passed the high road between Yü-lin Fu and Pao-t’ou, a large town on the northern border of the Ordos, The fort consists of three massive blocks of masonry built one upon the other. The largest at the base measures ninety feet square, while the whole edifice is about ninety feet high.

The Wall all along the Ordos border has crumbled away, leaving little more than a slight ridge. The watch towers which are situated at intervals of about three hundred yards, are, however, intact and serve still to mark the boundary line. It is obvious that the Wall in these parts was not faced with brick.

Yü-lin Fu is a prosperous military town with some seven thousand inhabitants. It is supposed to maintain a garrison of one thousand men to overawe the Mongols. Its walls are in good repair, and enclose many elegant temples. Like all the other border towns that we passed Yü-lin Fu is for ever threatened with inundation by the vast sea of sand that rolls in from the north and west. But for the river that flows past its western wall the city would long ago have been submerged, and even now the officials are continually perplexed with the problem, as sand, sweeping round from the north, banks itself against the eastern wall in many places right up to the battlements, forming an easy entrance for the belated traveller. We explored all the temples inside the city, but found nothing of extraordinary interest.

At length having made a sufficiently large collection, we decided to make a move, and on May 14 left the city by the East-Gate, heading in a north-easterly direction. Gradually we approached the Wall, and did not lose sight of its grey towers till we were within a day’s journey of the Yellow River and Shansi.

At intervals of about twelve miles we passed the dilapidated remains of small garrison towns now inhabited by a few poverty-stricken families. The country as we journeyed became more broken, and the mules had considerable difficulty in negotiating the steep slopes covered with drifts of sand. Towards the end of the first day we found ourselves on somewhat more elevated land, cut up in every direction by frightful chasms, which descended abruptly for hundreds of feet. Here our mules were momentarily threatened with destruction, as the path frequently led along the very edges of these precipices, and the loose soil continually broke away under their feet.

These chasms have certainly formed since the building of the Wall, for they frequently cut right across its length. To circumvent an unusually long one, we had to cross the Wall and make a long detour into the desert.

PLATE III.

[_Photo by M. P. Anderson._

IDOLS IN TEMPLE CAVE NEAR YEN-AN FU.

_Opposite p. 30._]
]

That night we put up at an isolated inn called Tuan-ch’iang, situated beneath the shelter of the Wall, and on the edge of a chasm some five hundred feet deep.

The following day another long detour into the desert was made, and we stopped for our midday meal at a little village situated at the end of a ravine in the sand. A clear stream, springing from beneath the wall of sand at the end of the ravine flowed along its bottom irrigating some small fields of black clayey soil, which supported good crops.

We followed down the stream till we reached a river, along whose banks we travelled till we came to a fair-sized town named Kao-chia-pu, where we put up for the night.

A guide, whom we engaged, here told me that forty years ago, when he was a boy, the Mohammedan rebels crossed the Ordos from Ning-hsia in the west, and devastated the whole region. He distinctly remembered their entry into Kao-chia-pu when they massacred all who did not escape into the surrounding hills.

The following day after a long hard march of some thirty miles we reached a large town named Shên-mu Hsien. We had now left behind us sand and loess, and were in hilly country composed mainly of sandstone.

On May 19 we descended from the high ground, and travelled down a long valley, making our midday halt at a small place named Ku-shan-pu.

My companion and I were some distance ahead of our mules when we entered the village, where a fair was in progress. Finding an inn we sought our usual rest from the noonday heat, but were soon disturbed by the crowd which became very aggressive.

I went out to pacify them and before I was aware of what was going on, found myself the centre of an angry mob. Cries of “Sa! Sa!” (kill, kill) assailed my ears, and the excited natives began to stone me. Convinced that my end had come, I resolved to sell my life dearly, but had no weapon save my bare fists with which to defend myself. I struck out right and left with all the force of desperation, feeling a grim pleasure as each successive blow laid one of my assailants low. At last my chance came. I was fighting towards a wall so as to gain protection for my back, which was being badly bruised, when the ringleader of the gang of desperadoes came for a moment between me and my goal. With all the force I could muster I let fly at his head, driving it with a thud against the wall. Down he went like a pole-axed ox and lay stunned. At this the mob broke up and fled. By now my temper was thoroughly roused and I resolved to teach those cowardly outlaws a lesson. Chasing the flying mob, I jerked two or three over onto their backs by sharp tugs at their outstretched queues. One man in his haste to get away tumbled head foremost into a large earthen jar of water. The mob made a rush for the gate of the inn yard, where they were met by Anderson, who was returning, all aboil, from a sharp fight in the street. There in the gate we caught the twenty odd, now cringing, Chinese, and hammered them, till, in sheer desperation, they pushed past Anderson and fled up the street. We next cleared the street and then bruised, with bleeding knuckles, and torn clothes, but triumphant, we returned to the inn to secure our interrupted rest.

The end of the day found us at Fu-ku Hsien, on the banks of the Yellow River, and crossing in a ferry we put up at an inn in the suburb of Pao-tê Chow.

These two towns are situated on high loess hills facing one another on either side of the Yellow River, the passage of which they are supposed to hold against invading Mongols.

Neither of the towns is populous, but large suburbs have sprung up on the banks of the river below them.

At Pao-tê Chow we stayed for a few days collecting, but meeting with little success, we left the place and headed for a wooded area in the mountains some two days’ journey to the south-east. Here we secured quarters at a farmstead situated at the top of a high loess ridge.

The surrounding country was well wooded and we made excellent collections of voles, mice and hamsters. Two new voles (_Microtus johannes_ and _M. Inez_) were secured in this district. Some vipers (_Ancistrodon intermedius_) were also caught.

The forest and woods, which were rapidly being cut down, consisted mainly of spruce, larch and pine. The open hillsides at the time of our visit were ablaze with the rich yellow blossoms of the wild rose and the delicate mauve of the lilac. Some varieties of hawthorn and blackthorn were also noticed, while a species of wild pear (_Pyrus sp._) was found growing on the hilltops. In places the ground was thickly covered with Solomon’s seal, lilies and other flowering herbs.

The mountains themselves were composed of limestone, of the usual grey colour, covered in most places with thick deposits of loess and a very fertile soil. In spite of the abundant vegetation the district was very dry and drinking-water scarce. Daily our mules and those of our host had to make a six mile journey to secure water at a spring which was rapidly drying up, owing to drought, and the number of farms it supplied.

After a stay of ten days we moved on to K’o-lan Chou, a dilapidated town some three days’ journey north-west of T’ai-yüan Fu. From here we travelled northward to Ning-wu Fu, passing Wu-tsai Hsien and Ch’ên-ch’êh Hsien on our way, both of which towns are situated on a wide plain stretching away to the Yellow River and Ordos Desert. Neither of these towns is of much importance.

Ning-wu Fu is a poor looking city of no great importance. Its walls, temples and gate towers are in a sad state of disrepair. It is situated on the Hui Ho, a tributary of the San-kan Ho, near the source of the Fên Ho. The inner loop of the Great Wall passes from east to west about five miles to the north, and its towers may be seen from the city wall.

On a hill behind the city lie two very large tower-forts, long abandoned and allowed to crumble and decay.

The business quarter of the city is small and unimportant, showing how poor the inhabitants are. There is no special produce or manufacture, and but for the fact that the surrounding districts are governed from this city, it would doubtless cease to exist as such.

At Tung-tsai near Ning-wu Fu we collected a good series of molerats (_Myospalax fontanus_), chipmunks (_Eutamias asiaticus intercessor_) and pikas (_Ochotona bedfordi_). Of these the molerat, which was first identified as _Myospalax fontanieri_ was new to science, while the chipmunk proving to be an intermediate form between the Chihli species (_E. a. senescens_) and that from the Ordos (_E. a. ordinalis_) was described as a new subspecies.

Single specimens of the little wood pika (_Ochotona sorella_) and the allactaga (_Allactaga mongolica_) were secured. The former of these turned out to be a new species, and it is very rare.

On June 26 we left Ning-wu Fu, travelling in an easterly direction as far as a small town named T’ai-niu-tien, after spending the night at which place we continued as far as Hsin Chow in a south-easterly direction. From here two easy stages brought us to T’ai-yüan Fu on June 30.

We had been away for five months, during which time we had not set eyes on a European other than ourselves, had completed a circuit of considerably over one thousand miles, and had made a collection of three hundred and fifty mammals including thirty-three species, twelve of which were subsequently described as new. Some dozen species of reptiles and batrachians were also secured, besides a small collection of birds.