The sides of the mass presented to view in the accompanying sketch* are evident fractures caused by the breaking up of the field by the deepening of the river in its course, and although the present mass is of the largest dimensions met with of this formation, I have no doubt it forms but a mere particle of the field as it originally stood, the remains of which now occupy various isolated positions in the river abreast of it, which run across two-thirds of the whole breadth.

Many other observations might be made on this deposit (and this specimen of it in particular), that do not now occur to me; but they will readily suggest themselves to some more intelligent visitor, who may be induced, from these remarks, on passing the spot, to give it an hour's examination.

As I believe no instance is on record of any other organic remains than shells having been found in those strata of kankar opened in so many parts of the Dúab, in excavating wells, and for the purpose of being burnt into lime, &c., the conclusion I draw from the observations I have been enabled to make, are all in favour of the opinion given in my letter of the 2nd of August, that I do not consider the fossil remains of the Jumna, as at all connected with the natural kankar formation, for wherever the specimens hitherto collected have been found, circumstances quite as conclusive as those above pointed out attend to shew that only these mechanically formed masses are in the slightest degree connected with the fossils, and that the formation is decidedly confined within the action and limits of the river, either past or present; but very possibly similar ones may be met with in parallel situations in other parts of the Dúab, generally alluvium, as yet unrecorded.

In your note on the Narsingpur fossils, I consider A A, the rocks in which the bones are imbedded, to be a most accurate description of the deposit rocks in the Jumna, if kankar was substituted for rounded pebbles of course, this difference the localities of these specimens has alone effected, as the distance from the hills (which alone could supply rounded pebbles of the Nerbadda, at Narsingpur) is so much less than the Jumna at Pachkowrit.

:

* A rough pencil sketch is here given in the MS. of the mass of kankar "of the deposit formation," containing the fossil elephant near Pachkowri; it lies 44 feet above water-mark: the description in the text has been deemed sufficient without the plate.-ED.

†This conglomerate varies its character according to the rocks which have supplied the rounded pebbles of which it is composed; these are sometimes granite, sometimes kankar, and sometimes jasper or vitrified clay.-A description of extensive deposits of it in the Rajmahal hills will be found in the extract from

M M

The position too of the rocks shewn in section, (fig. 1, Pl. 21, of Vol. II.) as containing fossils, is such, as I should have given them, had an elevation of the bank of the Jumna been required of me. Of course, I have had no opportunity of comparing the specimens from the above places; but from their general coincidence in position, and the fossil remains found in each, I am led to believe an intimate connexion exists between them in date, formation, and structure, and if, Sir, you think I have satisfactorily shewn the system of the deposit kankar formation in the Jumna, I think the same description would apply to similar formations in the Nerbadda.

3rdly. Of the Kankar Shoals.

These are composed of every variety of substance that is ever in motion in the Jumna, the most common of which are broken bricks, bones, shreds of earthen vessels, wood, fragments of granite, sandstone, quartz, agate, water pebbles, petrified clay, and composition shingle, of every variety of mixture that the clay of the surrounding country and sand of the Jumna will admit of. This last bears a proportion of four-fifths to the whole, which being mistaken for kankar, (of which the quantity is very trifling,) has occasioned the misnomer of kankar shoals.

It is among this heterogeneous assemblage of substances, that the best specimens of petrifaction are to be found. Bones, however, in every stage between freshness and a state approaching the hardest stone are procurable by turning over the surface about a foot deep; but I imagine, in fact I have ascertained, that not only more perfect, but a considerable abundance of the best specimens would be found at greater depths; as, during levels of the river sufficiently high to cover these shoals, the fragments near the surface are subject to violent attrition, and bones and other fragile substances, to total demolition, from the masses which are at such times continually rolling over them. Numerous instances occur in some of these shoals to support BUCHANAN'S MSS. published in the GLEANINGS, vol. iii., where also its characteristic of containing "giants' bones" is preserved in the very name of the place, Asurhár :—this circumstance has been brought to our notice lately by Mr. STEPHENSON, who has lately learnt that a gentleman at the Burdwan colliery has collected a number of fossil bones, and shells from the sides of other hills of the same range. Being very anxious that this field should be again and more thoroughly explored, we have republished the passage from Dr. BUCHANAN on the cover of the present No., and would direct the particular attention of our correspondents at Monghyr, and of the engineers engaged on the Rájmahal canał survey, to the whole line, which will probably prove as prolific as the Nerbadda or the Jumna. It may also afford proof against Mr. DEAN'S account of the formation of the conglomerate, and introduction of the bones within it by the action of the river.-ED.

the opinion before advanced, namely, that the force of heat is capable of causing the decomposition of bones, unless shielded by some intervening substance, applied during a state of freshness, and continued up to a certain period, the time of which must vary according to the quality of the bone; but my experience does not enable me to set bounds to the time necessary to render one of any quality proof to the effects of the sun's rays. I imagine, the seasons may cause so much variation, that the exact time necessary for them to continue under this protection, cannot be better defined than between the time of their deposition in a state of freshness, and the extinction of every animal or vegetable property, when they become nothing more than consolidated earth; (see specimens, the remainder of a pípal tree, Nos. A 3,) and even in this state I am led to believe, that exposure to the sun would cause decomposition, and to this, as well as to the effects of attrition, must be attributed the very few perfect bones found in these positions. I once found the femur of a camel, the middle of which was covered by a large damp stone, the portion covered was perfectly petrified in its whole circumference, whilst both ends were decomposed; but the absence of fossil remains in the whole section of this general alluvium is more conclusive than any minor proofs that can be adduced.

Very few specimens of wood occur in these situations. To the reasons advanced in explanation of the imperfect state of the bones is to be added the greater degree of brittleness of this substance in a petrified state. I have never procured more than three specimens from the kankar shoals, which I will forward with the other specimens of the collection.

Petrified clay (Specs. 9,) is found generally in small portions, and is transmuted by the same process as the earthy substance, to which wood is reduced previous to petrifaction, (spec. A 3,) which to all appearance has every property of indurated clay, the specific gravity of each being nearly the same.

Composition shingle, or cement pebbles, are produced by the admixture of clay or sand in almost every proportion of each: the most common process of the formation is as follows:

After a heavy shower, the water in its passage through the ravines near the river brings down with it clay in the shape of a thick sediment; this in many instances, after leaving the mouth of the ravine, has to run over large sand beds before it reaches the river, through which any considerable body of water cuts deep passages or gulleys, which run nearly horizontal 10 to 20 yards, and then fall 4 to 12 feet; running on again, they fall and run on irregularly, until reaching the

river. When the principal body is passed, the sediment becomes thicker, and dropping over these falls, mixes with the sand of the horizontal run beneath, forming first a single irregular mass on the upper side, whilst the under is pretty irregular, and of a rounded form: in this at first the sand predominates, the sediment continues dropping and adding to the stone, until all the sand within reach has been sucked in, when the formation ceases, and all the sediment that continues to fall on the same spot, adds nothing to, but merely rests on the composition, and is washed off by the next run of water, leaving a perfect stone. Six or eight stones are very frequently formed in this manner, of different shapes and varieties of composition, under the same fall, which is entirely regulated by accident; in some of these sand predominates, (specs. 10 and 11,) in others clay again, the composition consists of nearly equal portions of each. One fall may produce 10 or 12 stones separate, which another run of water may from the sediment falling on a layer of sand deposited since their formation unite, thus forming one stone, (spec. 12,) the difference between the first formed and their cement being very perceptible. The cement becomes set and as hard as dry mortar in two hours after the mixture has taken place, and after three days' exposure to the sun, they attain the substance of stone more or less hard, according to the justness of the proportion of the composition; these stones being generally round, are more frequently in motion than any other substance, and is owing to mistaking them for natural kankar, (I say natural, as I believe the substance to be kankar, of mechanical formation, the same ingredients forming in my opinion both,) that the term kankar shoals has been applied.

Of the Sunken Trees.

This dangerous obstacle to navigation is so well known from its occurrence in almost all navigable rivers, whose banks are covered with wood, that little need be said of it here.

The trees have originally occupied a position on the verge of the bank, which the stream having undermined, they have fallen into the river, with a quantity of earth attached to the roots, the weight of which firmly anchors them to the bottom, the head laying with the stream. In the Jamna any portion visible above the lowest levels is cut off to the water's edge by the inhabitants of the nearest village, leaving the bluff stumps of the large branches in the most dangerous position possible, at average levels. In 1833, the whole of these between Agra and Allahabad were sought for and taken out, and by the precautions then taken by the superintendant, it is next to impossible that any other instances can occur for many years, as every tree with

in a certain distance of the river has been cut down, and others still farther back marked for the same purpose 10 or 12 years hence.

A few may perhaps be drifted out of the Chambul and other tributary streams, but of so little consequence from their small size (the large and dangerous ones lying where they fall), that this obstacle may be said to be almost entirely surmounted.

Description of a cluster of four palms and a pípal tree. These remains have belonged to trees once growing on the general level of the Bundlecund bank, which having been sapped by the stream, they have slipped down with the earth, in which they grew, in the manner represented in sketch No. 7, (7. fig. 1. Pl.)* The pípal having been nearest the river has fallen lowest, and according to their distance from the edge do they now occupy their present positions, forming as it were a graduated scale, proving more strongly than any other instance I am aware, the petrifying qualities of the water. All I could write on this subject would not be so conclusive of this assertion as one glance at the specimens, which I shall merely describe.

A is the bark of the pípal stump, five feet in diameter, and about -14 feet long, lying on a sloping bank, with the root towards the river.

A 3 are portions of the body or trunk, which is reduced to that state, which I conceive necessary for any substance to attain before petrifaction commences, viz. a total extinction of all its animal or vegetable properties: whether the wood is actually changed into stone, or the gradual formation of stone merely destroys and takes its place, I am not able to decide; I can only say, when once properly petrified, the rings, the marks of annual growth of the tree, remain as apparent as when in a vegetable state.

A 2 roots of do. in a similar state to A 1.

B remains of the palm No. 1

The very apparent difference of texture between specimens Nos. 1 and 4, is caused by the former lying lower; it has been more frequently exposed to the action of the water than the latter. Nos. 2 and 3, occupy intermediate levels.

In adopting the term "petrified," as regards the palms, it is necessary to observe, that the striated fracture precludes the idea that this is the wood, the grain of which would be longitudinal, and confirms it as a tufa formation, enveloping the several parts of the tree exposed. Still I imagine, there are sufficient portions of fibres really petrified, to warrant its being applied as a general term to these specimens. * We have conceived it unnecessary to insert this sketch.-ED.