CHAPTER VIII.

THE HISTORY OF A RIVER.

IN the Introductory Chapter we learned, from a simple observation, that rivers derive their supply of water from the rain which falls on the surface of the land; and from the descriptive survey of the phenomena of snow and glaciers in the foregoing chapter, we have further learned, that an additional source of supply is the snow which falls on the summits of high mountains and on the plains of cold countries; and which, after remaining for a longer or shorter time in the solid form, is melted under the rays of the sun, or by the warm winds that set in with the returning spring. We must now enter somewhat more into the details of this subject, noticing the different modes, direct and indirect, by which the waters thus provided are gradually collected in the channels which carry them to the sea; and also the work performed by rivers in excavating these channels, and in wearing down and shaping the general surface of the land.

For our first example, we will take the case of the river Mahánadi, (any other large river of the peninsula would do as well,) a river, no feeder of which comes from a cold and snowy region, and which is, therefore, supplied solely by rain and the waters which issue from the earth in the form of springs. Take any good map of the country, (of preference, one of those skeleton maps issued by the

CHAP. VIII.] CATCHMENT BASIN AND DELTA.

105

Surveyor-General's Office in Calcutta, on which, being unencumbered by the names of places, the course of the rivers can be the more clearly seen), and observe the ramifications of this river and its tributaries. The general arrangement, as laid down in the map, reminds us somewhat of a tree or shrub stripped of its leaves, and with the roots partly drawn out of the ground. Only at one part of its course, and that for a short distance, are all its waters contained in a single channel, which we may take to represent the trunk of our ideal tree. This is above the town of Cuttack, where the river issues from the hill country. Above this, the main stream may be traced far up into the country, but becoming less and less the further we follow it inland. On both sides, it receives numerous tributaries from the country around; and if any one of these, the Til, for instance, be followed up in like manner, we see that, like the main stream, it is continually receiving minor subsidiary branches; the whole being not unlike the twigs and smaller branches on the bough of a tree. Observe, too, that when once they have united to form one stream or river, they do not again separate; but continue as a single stream, until the contents of the whole are collected in the main river, some distance above the point at which we commenced our survey. Now let us trace it downwards. At the point where Cuttack is built, it divides into two main branches, and each of these in its turn gives off others; the main channels becoming smaller and smaller as they approach the sea. It is something like a reversal of the former picture. Something, but not quite: for the subordinate channels into which the main stream branches out do not in all cases preserve an independent course. Many of them after a time reunite, so that the whole system may be compared rather to the meshes of a net than the ramifications of a tree; and the channels that finally reach the sea, though numerous, are not so numerous as they are a little way inland. We thus recognize a marked distinction between the upper and lower portions of the river. In the one, all the streams converge or run together; and the

106

FORMS OF VALLEYS.

[CHAP. waters from all parts of the country are collected step by step and united in one main channel. In the other, the channels diverge or rather reticulate, the waters being distributed between them: and when the river is unusually full, part of the water may even leave these channels and spread far and wide over the country. These two tracts are therefore distinguished as the collecting ground or catchment basin and the alluvial tract or delta of the river. The first is almost always much the larger.

These differences of the river's character depend on, and at the same time affect, the form of the land surface. Above Cuttack the country is either very hilly, or where, as in Chhatisgarh, hills are absent, gently undulating,-never absolutely flat; and every river flows at the bottom of a long depression termed its valley. Among hills, this valley is generally very narrow, and bounded by steep, sometimes rocky sides. Such is the case for instance at Neuraj, a few miles above Cuttack. At other places, the hills recede on either side, and the land surface slopes gently down to the river channel; but excepting perhaps a narrow strip of flat meadow land, along one or both banks, which may be covered by the river in high floods, the whole surface is permanently above the flood level. All the rain that falls on the country, in excess of that which is absorbed by the ground, or evaporated by the winds, runs down into the nearest streamlet, and contributes its quota to the volume of the river.

The river is its own engineer. The channelled depression, in which each stream flows, is the work of its own waters; and the engineer who cuts his canal with gently sloping banks in the soft soil of alluvial land, and with steep walls through a hard rocky barrier, does but imitate unconsciously the example set before him by nature. But while the former works with foresight and intelligence, selecting that course which appears to him on the whole the easiest, and to require the least expenditure of labour; sometimes cutting through a hard stubborn barrier in his path, with a view to shorten the distance, and to attain some advantage on the further side ;

VIII.]

WATERSHED OF MAHÁNADI.

107

the river, acting under the influence of gravity alone,' can only take that course which is for the moment the easiest,that by which it immediately reaches a lower level; and sometimes it makes a long circuit around some small obstacle, a hillock or low ridge for instance, which it cannot surmount, because an outlet exists at a distance at some lower level, through which the waters must escape. Hence it follows that rivers always follow a more or less circuitous course. Of this, the Barák river in Kachár offers a striking example on a small scale, and on a larger scale the Indus and Bráhmapútra. The drainage basin of every river is bounded by a line termed the watershed, or water-parting, which separates it from the adjoining river basins. This is frequently the highest line of a range of hills or mountains; but not always. Sometimes it is merely the most elevated part of a great plain, such as that which separates the drainage of the Jumna from that of the Sutlej to the west of Delhi; but this is the exception rather than the rule. As an example, we will trace out the watershed of the Mahánadi. Starting from a point on its north bank opposite the town of Cuttack, the line dividing the drainage of our river from that of the Bráhmani runs in a westerly direction, through the hilly Maháls of Atgarh, Ongol and Radakol. Hence it strikes northwards through a country of much the same character, till it reaches the high plateau of Chutia Nágpúr; separating the Ib, a large tributary of the Mahánadi, from the Brahmani. Both these streams take their rise in this plateau. From this point, we follow it to the west, along the crest or highest continuous line of the plateau, to the north of which the drainage runs into the Són river. After descending to a lower level, at the termination of this plateau in Sirgúja, the watershed makes a considerable détour to the north around the sources of the Hasto, in the plain of South Riwa; and then, returning to the south, it ascends the great hill mass of Amarkantak, where three great river basins meet. These are the basins of the Són flowing into the Ganges, the Nerbudda (Narmada) into the Gulf of Cambay, and

I See Introduction.

108

RAINFALL OF MAHANADI BASIN.

[CHAP.

the Mahánadi. Hence the watershed strikes southwards along the Mandla hills, the western slopes of which drain either into the Nerbudda, or the Wainganga, a feeder of the Godávari ; then, crossing the plain to the south-east, which lies between Nágpúr and Chhatisgarh, it enters the wild rugged hill country on the borders of Bastár, in which the main stream of the Mahánadi takes its rise and finally, running to south-east through the hills drained by the Tíl river, the largest tributary of the Mahánadi, it turns to north-east through the hills that separate the drainage of the Tíl from that of the smaller streams of the east coast; and then again to east, following a line parallel to the main river, back to our starting-point at Cuttack.

:

The country comprised in this long circuit is 45,000 square miles in extent. At least three-fourths of this is covered with rugged hills, for the most part rocky or clothed with forest; and the only cultivated tract of large extent is the fertile plain of Chhatisgarh. These characters of the basin, together with its form, which is round and compact with a narrow neck, determine certain peculiarities in the river, especially its liability to become flooded; but before noticing these, we must say something of the rainfall by which it is fed.

Most of the rain that falls on the Mahánadi basin, is brought by westerly winds from the Arabian Sea; and these winds prevail from June to September. In the hot weather months, a few showers come from the Bay of Bengal; and in October, at the close of the south-west monsoon, rain from the same quarter sometimes extends far up into the country. But the heaviest rain is that which falls during the south-west monsoon; brought, as I have said, by westerly winds. If all the rain that falls during a year over the whole basin could be collected, without loss of any, on a perfectly level surface 45,000 square miles in extent, it would cover it with a sheet of water about four feet, or perhaps fifty inches deep. This fact is usually expressed by saying, that the average annual rainfall of this river basin is fifty inches. All this water, however, is not carried away by the river. A part of it, we scarcely know how much, is re-evaporated, and passes