1. These two appeals arise out of suits brought by the proprietors of the Gopalpur estate to establish their right to two lankas situated in the bed of the Vasishta branch of the river Godavari. The major appeal A.S. No. 58 of 1942 relates to what is alleged to be a re-formation of or accretion to, the Ralaramuni lanka adiacent to the zamindari villaee of Udumudi. This lanka lies between the seventh furlong of the 24th male and the fifth furlong of the 25th mile according to the mileage on the left bank of the river, and it is about thirty miles from the sea. A.S. No. 441 of 142 relates to the Beillamoudi lanka situated between the 28th mile and the sixth furlong of the 28th mile that is to sa v. lower down the river. The evidence is to a large extent common in both anneals and the general questions arising are also common. I will deal first with the major appeal A.S. No. 58 of 1942 and thereafter consider the special features of the Ballampudi case.
2. The village of Udumudi belonging to the Gopalpur estate is situated on the left bank of the Vasishta Godavari more or less opposite to two Government villages the northernmost being Nadupudi and the southernmost Pedamallam. The river flows roughly north to south. The mileage along the banks of the river is reckoned from the Dowleswaram anicut which goes right across the river Go-davari just before the various branches of the delta divide. The land claimed by the Gopalpur estate consists of a roughly triangular island with its apex in the south-eastern comer and its eastern side running more or less parallel to and within a furlong of the eastern bank of the river. This island consists of a central portion which is now under cultivation and an outer fringe of sand planted with nanal or reeds the usual procedure in reclaiming a sand-bank from the river being to plant reeds to assist the deposit of fertile silt.
3. The island claimed by the plaintiffs lies roughly north of the site of the Balaramuni lanka as located in the earliest plan we have on record, namely, the survey of 1863. To the north of the site now in dispute there was in 1863 another lanka called Merka lanka. In 1863 the northern' portion of the Balaramuni lanka belonged admittedly to the Gopalpur estate except for a small strip on the western side which was attached to the Government village of Nadupudi situated on the western bank of the river. The whole of the southern portion of this Balaramuni lanka was in 1863 attached to the Government village of Pedamallam which is also situated on the west bank of the river. Similarly with the Merka lanka the southern portion of the lanka was attached to the estate village of Udumudi but there was a strip to the west of this which appertained to the Government village on the western bank. We do not know on what principle this division of the lankas, which lie almost entirely in the eastern half of the river, was made. The original grant of the Gopalpur estate is not produced. The survey of 1863 does not indicate that any portion of the bed of the river, apart from these islands was attached to the Gopalpur estate. The position, therefore, in 1863 was that both to the north and to the south of the area now claimed to be part of the Gopalpur estate there were islands owned partly by the Government and partly by the estate, the portions belonging to the estate being those portions adjacent to the area now claimed. It may be mentioned that the extreme northern portion of the area now claimed actually overlaps to a small extent the southern extremity of the Udumudi Merka lanka belonging to the Gopalpur estate as it existed in 1863.
4. We know nothingabout the state of the river between 1863 and 1878. In 1878 however the Gopalpur estate was under the Court of Wards and a survey was made of the lankas belonging to that estate. There was also a land regis-ter prepared and with the help of the rough plans and the register the dimensions and shape of the lankas and the position of the individual survey numbers therein can be fairly accurately fixed. Exs. A series are the plans prepared in the Court of Wards survey. They are not drawn to scale; but the measurements, including a limited number of diagonal measurements, are given, so that it has been possible to plot this survey to scale. Ex. MMM is a plan drawn to scale of the Balaramuni lanka as it existed in 1878. It is clear from this plan that the Balaramuni lanka had extended in a northerly and easterly direction so as to comprise nearly double the area which fell within the Udumudi village in 1863. There was also an extension of the Merka lanka in a southerly and easterly direction. The effect of the accretions to these two lankas in relation to the present suit claim is clearly illustrated by a combined plan which has been prepared by the Lankas Deputy Surveyor at the instance of the Court during the hearing of this appeal. This plan has been exhibited in appeal as Ex. XXXIX. It will be seen from this plan that whereas no part of the Balaramuni lanka in 1863 overlapped the site now claimed by the Gopalpur estate, a considerable portion of the accretions to this lanka overlaps the south-west of the site now claimed. Similarly, the accretion to the Merka lanka at the time of the Court of Wards survey of 1878 extends very considerably the area of this lanka which coincides with the northern portion of the area claimed by the plaintiffs. Without taking actual measurements it would appear that between one-third and one-half of the site now in suit coincides either with the Balaramuni lanka or with the Merka lanka as they existed in 1878. It seems to me that there can be no question regarding the title of the estate to the accretions surveyed in 1878 as part of these two lankas falling within the estate. No doubt the survey may not strictly bind the Government, though it was presumably conducted by a Government Surveyor at the instance of the Estate Collector. But clearly the results of this survey were embodied without question in the land registers of the estate and it can be demonstrated that for many years the estate without objection from the Government treated the accretions to the two lankas as its own property, collecting revenue and auctioning those portions of the lankas which were not given on patta.
5. There is no plan of the river between 1878 and 1895, but the estate registers, Exs. DDD-14, UU, Z-l and C show the stages by which the northern and eastern portions of the Balaramuni lanka were washed away in successive floods. By the time of the survey of 1895 commonly known as the 'Water Tax or Mamool Wet Survey' embodied in the plan Ex. II, the actual extent of the Balaramuni lanka had fallen from approximately 119 acres, which was the extent in 1878, to a little over 38 acres. By 1895 the strip on the west of the lanka belonging to the Government village of Nadupudi had disappeared into the river. We are not of course concerned with the southern portion of the lanka which belonged to the Government village Pedamallam. The next plan is the River Chart Survey undertaken by the Government of India in 1903, the map of which is Ex. IV and the corresponding register Ex. KKK. This plan shows that the portion of the Balaramuni lanka shown as cultivable land belonging to the Udumudi village had been still further reduced by erosion in the north-eastern corner, but a sand-bank had fomed extending a considerable distance to the east and the west of the lanka. In all these early plans there is a channel of the river separating the Balaramuni lanka both from the Merka lanka to the north and from the main bank of the river to the east and the width of this channel which had decreased between 1863 find 1878 showed a tendency to increase in the later period. By 1903 the cultivable extent of the Balaramuni lanka in Udumudi village had fallen to 24.34 acres.
6. After 1903 there is a gap in the evidence until 1920. An attempt has been made on behalf of the estate to contend that Ex. XVI, a plan which shows a large island more or less over the site now claimed by the plaintiffs, belongs to the year 1919. The evidence is not at all clear. The plan bears at its foot the engineer's signature with a date in 1920, but at the top of the plan there is the draftsman's signature of July 1929. The plan was treated in the lower Court as a plan of 1929. A comparison of the conformation of the land with the plans of the years preceding and succeeding 1919 makes it very unlikely that this was a plan of that date. There is no oral evidence to show when or how this plan was prepared. The learned Government Pleader has suggested that this is one of a series of lanka survey plans which were usually prepared by inserting in a skeleton plan drawn up at an earlier date by the Public Works Department details roughly demarcated as a result of the Lanka Inspector's visits from time to time. It seems most probable that this is the correct explanation of Ex. XVI and that the date of the Superintending Engineer's signature is the date of the preparation of the skeleton plan of the river, while the details regarding the formation of the laakas were inserted in 1929, the date of the draftsman's signature. It is, however, regrettable that this point was not made clear by evidence in the lower Court. We may take it for the purpose of this appeal that Ex. XVI is a plan prepared in 1929 and is therefore of little importance since the litigation started in 1931.
7. Ex. V is the Lanka Inspector's plan of 1920-21. It is common ground that at or about this time the whole of the cultivable soil of the Balaramuni lanka had been washed away. Ex. V shows nothing but an irregular sand-bank in the southern portion of the site formerly occupied by the Balaramuni lanka. Proceeding northwards there is a channel with a small island, then roughly in the middle of the site now claimed by the estate we find a large, roughly quadrilateral sand-bank extending east to west. This is separated from the remains of the Merka lanka and from the bank of the river by a channel which runs first to the cast, then southwards, cutting off the various sand-banks from the main river. Some doubt is thrown on this plan in so far as it shows a flowing channel between the remains of the Balaramuni lanka and the sand-bank to the north of it. The karnam has deposed that there is no water in this channel. It may well be that this was a shallow depression which would be dry during the low water months and would contain water at other times of the year. The fact, however, remains that the whole of the Udumudi Balaramuni lanka had been washed away except for the existence of a sand-bank, presumably submerged when the river was high, and there was nothing in the nature of cultivable land in the Central sandbank, part of which corresponds to the area now claimed by the plaintiffs.
8. The lanka now claimed by the plaintiffs seems to have begun to form as cultivable land somewhere about the year 1924. At any rate, Ex. XX, the survey of 1927, with the corresponding 'register Ex. VIII, shows a large survey number 1 with a sub-division 1|2 in which were patches of cultivation. This cultivation cannot be precisely located; but it would appear that the whole block overlaps part of what was originally the Merka lanka and part of what was originally the Bala-ramuni lanka; but it does not appear that there was at this time any channel flowing to the south of this block of cultivation. If Ex. XVI has been correctly dated in the year 1929, it would show that the block of cultivation by this time was in the centre of the island, that it was surrounded by nanat or reeds, and that the whole formation extended as far as the boundary between Pedamallam and Udumudi, there being a channel separating it on the north from the Merka lanka and on the east from the main river bank. Finally we have the plan attached to the encroachment notices which led to the suit showing an occupation roughly triangular in shape extending in a northerly direction so as to overlap slightly the original area of the Merka lanka, but extending in a south-easterly direction to a point some distance to the north of the boundary between Udumudi and Pedamallam and not overlapping the original Balaramuni lanka as it was in 1863.
9. It will be gathered from this summary of the documentary evidence that the area now claimed by the estate is a formation arising from the bed of the river at some time after 1920, assisted perhaps in its formation by the existence of sand-banks which may be the remains of the pre-existing lankas; but certainly it was not a lateral accretion to any existing land which was definitely above the normal high water mark at the time. It will be gathered also that this area now claimed by the plaintiffs lies partly over the site which was an accretion to the Balaramuni lanka in the year 1878, partly over a site which has always, so far as we know, been situated in the bed of the river, and partly over the site which in 1878 was attached to and treated as a cultivable portion of the Merka lanka in the north.
10. Now the claim of the estate to this formation is based on various contentions. Firstly, it is contended that the whole of the plot now in dispute was demarcated as part of the Gopalpur estate with field numbers 548 and 587 to 597.- This statement is not quite correct, for survey number 587 is shown in the Court of Wards survey as 'Godavari river'; and it is difficult to see how the action of the surveyor in giving to a portion of the river a number in the survey of the village describing it as 'river', could operate to grant the bed of the river to the estate which owned the village, in the absence of evidence that the original grant included any portion of the bed of the river. Similarly the plaintiffs contend that the whole of the disputed locality was surveyed as part of the estate in 1895 and that it has been recognised as part of the estate and cannot therefore be claimed by Government. It is contended that the disputed lanka has been in the continuous possession and enjoyment of the Gopalpur estate and that the estate is entitled to it either as an accretion to or a re-formation of land belonging to the estate. The plaintiffs also assert that the Vasishta Godavari river is not a tidal and navigable river in the disputed locality and that the bed of the river up to the middle of the stream belongs to the plaintiffs as owners of the land on the left bank. The plaintiffs therefore claim not merely the cultivated area, but also the sand shoals around it and ask for a declaration that the whole locality belongs to them and they also allege that even if the Government was once entitled to the land, it has lost it by adverse possession. The declaration given by the lower Court is that the plaintiffs are the owners of Balaramuni lanka of 93 acres 37 cents and also of about 100 acres of sand shoals surrounding it 'as per plan attached to the decree'. There is no plan attached to the decree amongst the records in Court and there arc no materials upon which the location of the sand shoals decreed to the plaintiffs can be precisely fixed.
11. The first question to be decided is whether the bed of the river belongs to the Government. We are not able to accept the contention of the learned Government Pleader that this question is governed by res judicata as a result of the decision in S.A. No. 785 of 1942 a case relating to a different portion of the same river. A great deal of time has been spent in the lower Court in considering the question whether the Vasishta Godavari at the disputed point is navigable throughout the year. The evidence seems to show that in times of low water, navigation is carried on with extreme difficulty by utilising the tide and making bunds and excavations in order to bring up boats capable of carrying cargoes. Light boats of very shallow draught such as are used by fishermen seem to ply through put the year. It seems to me, however, having regard to the known facts regarding the river that the question of navigability, if it arises at all, should not be considered with reference to conditions as they exist now but with reference to conditions as they existed at the time of the grant. If at the time of the grant the bed of the river cannot be deemed to have been conveyed to the Zamindar, any doubt regarding the navigability of the river in present condition would not operate, so as to convey half the bed of the river to the zamindar. In or about 1852 a great irrigation Work was constructed consisting of a dam or anicut right across the river at Dowleswaram, some twenty-five miles above the disputed locality, and an elaborate series of canals and channels, whereby water was intercepted and distributed throughout the delta. It is not disputed that the river Godavari above this dam is navigable throughout the year. The dam however intercepts the whole of the low level water which flows towards the sea after the floods have disappeared. The flood season is normally from the end of June to about the end of September. The extreme low water season is during the months of April to June. During the latter period the only water which flows in the Godavari river below the Dowleswaram anicut consists of percolation water, drainage water and such water as may be ascribed to springs. Seeing that the river is navigable throughout the year above the Dowleswaram anicut, it is reasonable to infer that before the anicut was constructed the main branches of the river, of which the Vasishta is one, would also have been navigable throughout the year.
12. The question whether the river is tidal at the disputed locality seems to me also fairly clear though I do not agree with the learned Subordinate Judge who has held that the river is not tidal. Ex. XXVI is a statement of tidal readings taken by the Public Works Department at the fifth furlong of the 26th mile during the month of June, 1937. This shows very considerable differences of level at different states of the tide. Thus on 6th June there was a rise and fall of nearly four feet. On the 13th June, the rise was three feet. In the middle of the month the differences are very slight. Towards the end of the month we again find differences of two to three feet. Quite clearly the Vasishta Godavari is subject to the influence of tides at this point which is roughly one mile below the area under consideration in A.S. No. 58 of 1942 and about two miles above the area under consideration in A.S. No. 441 of 1942. The Government expert D.W. 4 definitely states that the effects of the tide are noticeable up to Gopalpur, which is six miles above the suit locality. The plaintiff's expert, P.W. 25, admits that the river is tidal within twenty-five miles of the mouth and that if the bed has been scoured out the tide might come up about six miles further. The report of the Commissioner, Ex. GG, is to the effect that there is an ebb and flow of the tide at the disputed locality though the action is not sufficient to enable boats laden with cargoes to pass freely up and down the river in seasons of low water. On this state of the evidence it seems quite clear that the river at the point now in dispute is tidal and that it must be deemed to have been continuously navigable at the time of the grant to the plaintiffs. It follows therefore that the bed of the river belongs to the Crown.
13. This leads to the further question, what constitutes the bed of a river? This is a question which is not always easy to answer and two attempts to answer it have been frequently adopted by the decisions. In the case of Alabama v. Georgia 16 Lawyers Edition 556 at 560, the Supreme Court of the United States of America defined the bed of a river as 'that portion of its soil which is alternatively covered and left bare, as there may be an increase or diminution in the supply of water, and which is adequate to contain it at its average and mean stage during the entire year without reference to the extraordinary freshes of the winter or spring, or the extreme droughts of the summer or autumn'. This definition has been accepted by the English Courts as one possible definition for the guidance of the Courts, that is to say, the criterion is whether the land in question is situated
14. Lawyers Edition 556 at 560. within the area over which the river will flow in a normal condition and which will sometimes be covered and sometimes left bare according to the quantity of water coming down the river. Another criterion which is perhaps more applicable to conditions in India is suggested in the decision of Romer, J., in Hindson v. Ashby (1896) 1 Ch. 78. The learned Judge says 'I think that the question whether any particular piece of land is or is not to be held part of the bed of a river at any particular spot, at any particular time, is one of fact, often of considerable difficulty to be determined, not by any hard and fast rule, but by regarding all the material circumstances of the case, including the fluctuations to which the river, has been and is subject, the nature of the land, and its growths and its user.' The learned Judge quotes a decision of the Supreme Court of the United States where emphasis is laid upon the appearance of the soil, its vegetation and so on when deciding whether a particular piece of land is part of the bed of a river. This decision in Hindson v. Ashby (1896) 1 Ch. 78 came up in appeal [(1896) 2 Ch. 1] and the principles upon which the trial Court had proceeded were affirmed though the decision was reversed on facts.
15.Now applying these two tests to the present case we have to consider whether a mere sand-bank, visible when the water is low, submerged when the water is high, is to be regarded as part of the bed of a river or as an island capable of ownership by a person who is not the owner of the bed of the river. It is of course impossible to draw a hard and fast line between that which is a sand-bank still forming part of the river-bed and that which is an accretion or re-formation belonging to the owner of the land to which the accretion is attached or of the soil formerly existing at the point of re-formation. The process of formation in the river Godavari consists ordinarily in the rise of a sand-bank owing to the shifting of the river-bed by floods, the planting of the sand-bank with reeds, the interception of silt by the reeds grown on the sand and the gradual formation of cultivable land which will be considerably above the water level at all seasons except during high flood. It seems to me that until this last stage is reached, that is to say, until the soil has been deposited so as to raise the bank above the ordinary level of the water, there can be no question of any completed title by lateral accretion to the land to which the sand-bank is attached. If, however, the sandbank is a re-formation of land formerly owned and washed away, the owner may be presumed to have retained his title in the site of the land he formerly occupied and, when it re-appears, even though the re-formation is incomplete, he is entitled to a declaration of his right thereto, provided that the re-formation is identifiable with the original land and there has been no abandonment of his right therein. The leading cases cited before us on this subject are Lopez v. Muddun Mohan Thakoor (1870) 13 M.L.A. 467, Hurasahai Singh v. Syed Lootf Ali Khan (1874) L.R. 2 IA. 28, Maharajah of Dumraon v. Secretary of State for India in Council I.L.R. (1927) Pat. 481 and Rani Hemanta Debi v. The Secretary of State for India in Council and Sri Sundari Debi v. The Secretary of State for India in Council (1906) 3 Cri.L.J. 560.
16. There has been some discussion before us of the difficult theoretical question which arises when the bed of a river is increased by the gradual and imperceptible process of diluvion and subsequently land is re-formed on the site washed away. Will the accession to the bed of a river by this process of diluvion, which is the inverse process to that of accretion, prevail over the right of the former owner to retain Ownership of the site which was his at any rate until the river washed away the upper soil? I do not, however, find it necessary to propound an answer to this question for it seems to me that it does not arise on the facts of the present case if properly viewed. We know that portions of the site now in dispute were formerly occupied by the Merka lanka in the north and the Balaramuni lanka in the south, both of which were at these points recognised as appertaining to the Gopalpur estate. We know that the Balaramuni lanka was extensively eroded during the period from 1878 to 1903. There is, however, no material for holding that this erosion was a gradual and imperceptible process. In fact so far as the registers are available it would appear that extensive blocks of land were washed away in particular floods, that the extent so washed away was duly recorded and remissions of revenue were granted in respect thereof. In these circumstances it cannot in my opinion be said that the process of erosion was gradual and imperceptible so as to provide a basis for the argument that the title of the former owner to these sites so washed away has been extinguished.
17. Now summing up the position in the light of the evidence and of the law as summarised above, it seems to me clear that at the time when the suit lanka was formed there was no existing land of the Gopalpur estate to which it can be deemed to be a lateral accretion. Such sand-banks as remained appear to have been, so far as the evidence goes, mere elevations in the bed of the river not capable of effective occupation, at any rate in the years 1920-21. We are not therefore concerned with any question of lateral accretions. I may, however, observe that the attempt of the learned Government Pleader to dispute the title of the estate to the cultivable lands surveyed as part of the estate in 1878, on the ground that the process of gradual accretion is not established, must fail; for it seems to me that the recognition of the estate's right to these lands as accretions must have been undoubted, in view of the treatment of these lands both at the time of the survey and subsequently. The right of the estate to the cultivable lands demarcated as part of the estate in 1878 being established, it seems to me to make no difference whether the estate's right depended on an original grant or an accretion to that grant. The estate was the full owner of the cultivable lands shown in the plans and registers of 1878. Those lands were to a large extent washed away by a series of floods. There is nothing to show that the Government as the owner of the bed of the river acquired title to the portions so washed away. To the extent therefore to which these lands belonging in 1878 to the estate can be shown to have re-formed as part of the area now in dispute, clearly the title of the Gopalpur estate must prevail. To the extent to which the land now in dispute cannot be shown to be a re-formation of either Balaramuni lanka or the Merka lanka it must, in my opinion, be deemed to be a new formation in the bed of a river belonging to the owner of the bed, namely, the Government. The plan which has been prepared at our instance Ex. XXXIX shows the extent to which the two lankas as they existed in 1878 belonged to the plaintiffs' village of Udumudi and overlap the area now claimed by the plaintiffs. The title of the plaintiffs will be declared to the land and to the bed of the river included within the red lines marked on this plan and the plaintiffs will be entitled to a refund of so much of the penal assessment as has been levied with reference to land included within these red lines the amount to be determined by the lower Court. The appeal and the suit having succeeded only in part I think it proper that both parties should bear their own costs throughout.
18. I will turn now to the evidence regarding the Ballampudi lanka which is the subject-matter of A.S. No. 441 of 1942. The site claimed by the plaintiffs consists of 43 85 acres. The plaint definitely proceeds on the basis that this area is a re-formation of a lanka formerly belonging to the estate and identifiable with land that originally belonged thereto. It is also contended that if there be any, excess in area it would still be part and parcel of this re-formed lanka or that in any view of the case it would be a lateral accretion to the reformed lanka which belonged to the proprietors of the Gopalpur estate. There is no contention that the disputed land is a lateral accretion to the bank of the river belonging to the proprietors. An attempt has been made to argue the case on the footing that the land now-disputed was a lateral accretion to the strip of land bordering the bank which clearly belongs to the plaintiffs. On the facts I doubt very much whether this would be established; but it would be most unfair to require the defendant to meet such a case which is not, in my opinion, suggested either in the plaint or in the issues.
19. The Ballampudi lanka has a much shorter history, so far as documents go, than the Balaramuni lanka. The survey of 1863 shows no formation of any kind in the area in dispute and no accretions to the left bank of the river. The Court of Wards Survey in 1878 is also negative. At the time of the Mamool Wet Survey in 1895 there was still no lanka and the relevant survey numbers on the river bank are given as Nos. 88 to 90. In 1903 the River Chart Survey Exhibit VI shows two narrow strips Nos. 93 and 92, adhering to the riparian numbers corresponding to those already mentioned. It is conceded that these two numbers, 93 and 92, belong to the zamindar. To the west of these two numbers there was an extensive area, No. 89, described as 'Sand' with description 'dubbu grass' in the accounts. The accounts also show that originally the . word 'zamindar' was written against this number but was subsequently scored off. No creek is shown between this No. 89 and the two riparian accretions 92 and 93. There is no documentary evidence at all for the interval between 1903 and 1920. In 1920 the Lanka Inspector's survey shows a narrow strip of cultivated land adhering to the eastern bank, similar in shape and location to the two numbers 93 and 92 found in Exhibit VI. To the west of this narrow strip there is a creek which is shown as containing water for about half its length and below that there is marked what is described as a low paya blocked up. A paya is a channel, but not necessarily one which would contain water in the dry season. To the west of this creek and dry channel there is marked a sand shoal. The 1927 plan Exhibit XXVII shows a similar cultivated strip along the bank such as is found in the earlier plan. To the west of that in its northern portion is a formation on which were growing nanal and grass, below which there is shown a channel running east and then south along the edge of the cultivated strip cutting off another formation which is described as 'nanal or reed.' The application for permission to cultivate thirteen acres situated in the neighbourhood of the third, fourth, and fifth furlongs of the 28th mile is accompanied by a sketch Exhibit G-l. This cultivable area would correspond to the land separated in 1927 by a channel from the nanal growth in the river-bed proper. This sketch shows nanal in the space between the cultivated strip and the river-bed. In 1930 the Government issued an encroachment notice to which a plan Exhibit XXIII is attached. This plan shows an area of 37,35 acres in the occupation of the zamindar situated between the first furlong of the 28th mile and the sixth furlong of the 28th mile, separated by a silted up channel from the narrow strip adhering to the river bank. The documents prior to 1903 on which the lower Court seems to rely, Exhibits M, N, O, P and Q seem to have no reference whatever to the area now in dispute. They may well refer to the narrow strip adjoining the river bank to which the title of the zamindar is admitted.
20. The facts to be deduced from the plans are as follows:--Before 1903 there was no formation at all in the area under consideration. In 1903 there was ah extensive sand-bank on which there was grass, adjacent to the strip which the estate had brought under cultivation. There is no evidence that this sand-bank ever became an effective accretion to the land of the estate or that it was ever cultivated or brought under the occupation of the estate. We do not know anything about the later history of this area until after 1920 when a series of plans indicates the formation first of a sand shoal, then of land in process of reclamation and finally of a reclaimed area, separated from the strip adjoining the river bank by what was apparently always either 'a channel or silted up water-course. On these facts the only inference possible seems to be that the area claimed in this suit is a vertical formation, not identifiable as a re-formation of any land formerly belonging to the estate. As I have already stated, it is not claimed as an accretion to the bank, but I am of opinion that on the facts proved, even if such a plea could be taken, it would have to be found that this land has not been shown to be an accretion to the riparian land owned by the estate. The tidality of the river at this point is even less in doubt than that of the river higher up dealt with in the other appeal. Since this land must be regarded as a vertical formalion arising in the bed of the river and since the bed of the river belongs to the Government, the plaintiffs cannot claim this land. This appeal, A.S. No. 441 of 1942, is allowed with costs throughout and the suit is dismissed.
21. We fix the advocate's fees as Rs. 750 and Rs. 375 for junior in A.S. No. 58 and at Rs. 400 and Rs. 200 for the junior in Patanjali Sastri, J.--I agree. In view, however, of the importance of some of the questions argued before us I desire to add, a few observations. As my learned brother has narrated the relevant facts and reviewed the material evidence in his judgment which I have had the advantage of reading', it is unnecessary to go over the same ground here.
22. The larger lanka which is the subject-matter of A.S. No. 58 of 1942 is claimed by the respondents mainly on two grounds, namely, (1) as the Vasishta is neither tidal nor navigable at the locality in question, they, as riparian proprietors own the bed of the river ad medium filum aquae which includes the land in dispute and (2) even if the river is held to be tidal or navigable and the entire bed is in consequence vested in the Government, the appellant herein, the new formation is an accretion to or a re-formation in situ of certain old lankas admittedly appertaining to the respondents' Village Udumudi which lies on the eastern bank of the river subtending the formation. The appellant contests both these grounds and the question is whether they are well-founded.
23. The Court below has found that the river is neither tidal nor navigable at the point. But our examination ,of the evidence shows that it is tidal, though it is not navigable in its present state during the dry season (June to September) as the construction in or about 1852 of the anicut at Dowleshwaram 25 miles higher up diverts the waters of the stream in those months into a net-work of artificial canals and channels for irrigating the delta area. It is admitted on behalf of the respondents that, in spite of the construction, the seepage and drainage water flowing down the river below the dam is of sufficient volume and depth to render it navigable even during the summer for about twenty-five miles from the mouth of the river. The lanka now claimed is situated six miles higher up. It is not disputed that above the anicut the stream is navigable throughout the year. In these circumstances, it is argued for the appellant that, on the question of ownership of the bed of the river, tidality and not navigability is the determining factor, and that, even if navigability in the sense of being navigable throughout the year (see Secretary of State for India v. Venkatanarasimha Naidu (1919) 11 L.W. 256. and Subbarayudu v. Secretary of State for India : (1927)53MLJ868 .) were held to be the basis of the Crown's title, it must have reference to the natural state of the stream, and the cross-bunding cannot affect the character of the river as a navigable stream or the title of the Crown. The respondents, on the other hand, contend that in this country the ownership of the Crown in the bed of a river is deter-mind by its navigability throughout the year, and that the ebb and flow of the tide at the point is immaterial. In my opinion, neither of these contentions in its extreme form making tidality or navigability the sole test is correct.
24. The true position would seem to be that the Crown's ownership of the bed will be presumed if the river is either tidal or navigable. It is well established in English Common Law that the bed of tidal waters vests prima facie in the Crown. The reason of the rule is thus explained by Farnham on Waters and . Water Rights, Vol. 1, page 179:
The one unique feature of the sea is the tide. When, therefore it was established that the King had title to the sea, it was an easy step to the conclusion that the sea went as far as the tide did, so that the King's title extended up the rivers as far as the tide flowed;
and he cites the Banne case Davies Report 149 as saying that the, reason why the King has an interest in such navigable river as high as the sea ebbs and reflows is because such river partakes of the nature of the sea. Indeed, the title of the Crown to the bed has been held in England not to extend further than the tide though the river may, in fact, be navigable even above that point. In Hindson v. Ashby (1896) 2 Ch. 1 (9) Lindley, L.J., said:
There is no question here of any right upon the part of the Crown to the bed of (he river, for at Wraysbury the river, though navigable, is not and never was tidal. Hale, de jure Maris, Ch. iv.
25. Thus under the English Common Law the ownership of the Crown in the bed of a river ends and that of the riparian owners begins at the point where the stream ceases to be tidal, alike in navigable and non-navigable rivers. This limitation on the Crown's ownership of the bed has not always been accepted by the Courts in America where it is held in some cases that the bed of a navigable stream vests prima facie in the Crown even where it is non-tidal, such extension of the Crown's ownership being founded upon the necessity of preserving to the public the use of navigable waters from private interruption and encroachment. This view is vigorously criticised by Farnham who stresses the distinction between a public right and a public use, and points out that the public easement of navigation can well co-exist with the riparian ownership of the bed (Volume I, pages 247 and 248). The doctrine of the Crown's ownership of the bed of navigable but non-tidal rivers has, however, been recognised and given effect to by this Court in a recent decision in Maharaja of Pithapuram v. Secretary of State for India : (1942)1MLJ344 where the earlier cases have been reviewed. However that may be, there seems to be nothing in the doctrine that the ownership of tidal waters is prima facie in the Crown which precludes its application to this country, and no authority holding it to be inapplicable to the conditions here has been brought to our notice. Reference was made to the observations at p. 543 in the recent decision already referred to as supporting the view that navigability alone is the test of the Crown's ownership of the bed and tidality is immaterial. After referring to the Privy Council decision in Tarakdas Acharjee Chowdhury v. The Secretary of State for India in Council, the learned Chief Justice who delivered judgment of the Court observed:
Neither in this case nor in Haradas v. The Secretary of State for India in Council (1917) 26 G.L.J. 590 was any suggestion made that the tide is a factor in deciding the ownership of the bed of a public navigable river in India.
26. I do not read this passage as laying down that tidality can be no test at all in deciding the ownership of the beds of rivers in India. All that was apparently meant was that the tide was not an essential factor in determining title to the bed of a public navigable river in this country. The learned Judges did not have to consider the question of the ownership of the soil in the converse case of a tidal but non-navigable stream. I am of opinion that the title to the bed of the Vasishta at the locality in question where the tide ebbs and flows must be taken prima facie to belong to the Crown even though the river in its present state is not navigable there throughout the year.
27. The other branch of the appellant's contention also seems to me to be well-founded. Assuming navigability to be the sole basis of the Crown's ownership of the bed as suggested for the respondents, can it be said that the Vasishta is not navigable merely because the anicut intercepts almost the entire waters of the stream so as to render navigation difficult without artificial aids for some distance below during the summer months. The navigability of a river, I apprehend, must be judged with reference to its natural state, and if it was navigable before the construction of the dam and the title to' the bed was in the Crown, it is difficult to see how such title could be divested merely because the cross-bunding has reduced the volume of water flowing down the stream below the construction, any more than the title of a riparian owner could be divested by the river being made navigable by deepening the bed by artificial means. The Crown's title must be taken to continue unless the respondents are able to show that such title was lost by grant or adverse possession of the bed, and there is no room for applying the principle of riparian ownership extending up to the middle thread of the stream. No question of adverse possession was raised before us, and indeed no continuous possession of any portion of the bed could have been possible in such a fluctuating river which from time to time shifts its course encroaching, submerging and receding owing to the action of the mighty floods which flow down the stream. As for grant by the Crown, at the time of the permanent settlement of the respondents' estate in or about 1803, the river must undoubtedly have been navigable throughout the year at the locality in question, and there is no presumption that a grant of lands on the banks of a tidal or public navigable river passed title to the bed ad medium filum. The respondents have not produced their sannad which must be in their possession, and there is no satisfactory evidence to show that the river-bed where the suit lanka has sprung up was included in the respondents' estate.
28. Turning next to the other ground on which the respondents' claim was sought to be sustained before us, viz., accretion and re-formation in situ, the principle of accretion has thus been explained by their Lordships of the Judicial Committee in Lopes v. Muddun Mohun Thakoor (1870) 13 M.I.A. 467:
Where there is an acquisition of land from the sea or a river by gradual, slow, and imperceptible means, there, from the supposed necessity of the case, and the difficulty of having to determine, year by year, to whom an inch, or a foot, or a yard belongs, the accretion by alluvion is held to belong to the owner of the adjoining land.
29. In Secretary of State v. Raja of Vizianagaram , their Lordships observe that in dealing with the great rivers in India, it is necessary to bear in mind the comparative rapidity with which formations and additions take place and that the words 'slow and imperceptible' used in the earlier decisions are only qualifications of the word 'gradual' and must be understood and applied with reference to the conditions of the country where the question arises. In the present case there is no clear evidence to show that the new blocks of land shown in the survey maps Exs. A series of 1878 to the north of the Balaramuni lanka and to the south of the Merka lanka adjoining the portions which admittedly belonged to the respondents, were formed by a process which can be said *to be 'gradual' even in the qualified sense explained by their Lordships. These blocks of land are not to be found in the survey map of 1863, and as we have no reliable evidence regarding the state of these old lankas between 1863 and 1878 when the new blocks appear to have come into existence, it is not possible to say that their formation was so gradual as to entitle the respondents to claim them as accretions to their old Balaramuni and Merka lankas. The maps Exs. A series, the land Register Ex. B and the plan Ex. MMM relating to the survey of 1878, however, indicate that these blocks adjoining the old lankas on the north and on the south respectively were treated as integral parts of these lankas and were surveyed and demarcated as appertaining to the Udumudi village. As the survey operations of 1878 were admittedly conducted by the officers of the Government at the instance of the Court of Wards who were then in management of the Gopalpur Estate during the minority of the proprietor, it is not unreasonable to infer that the new formations shown in the survey maps of 1878 were treated by all concerned as increments to the respondents' lankas and as such belonging to them. During the subsequent period, however (i.e., from 1878 to 1895) we find that not only the new blocks referred to above but even considerable portions of the old lankas were washed away by the river, as the cultivation and other accounts of the estate show a steady dwindling of their areas. Here, again, it is not possible to say whether the irruption by the river was so gradual that the respondents could be said to have lost their title to the portions washed away on the principle of diluvion. Then followed a period during which the river began to throw up new formations again which, after various intermediate changes in shape and extent, assumed by 1930 the form of the island now in dispute. This block is separated from the Balaramsuni lanka on the south and from the Merka lanka on the north by payas or water-courses, and it does not adjoin the Udumudi village on the bank. It is thus clear that it cannot be regarded as a lateral accretion to any land belonging to the respondents, but must be regarded as an independent vertical formation in the bed of the stream. Title to it must therefore be in the Crown unless the respondents are able to establish by clear proof of identity that any part of it is a re-formation in situ of their old lanka lands washed away by the river.
30. The principle that land washed away but subsequently re-formed on its original site belongs to the original proprietor is well established (see Maharaja of Dumraon v. Secretary of State for India in Council (1896) 2 Ch. 1 (9)):. Whether this principle is applicable only where the land is lost to the original owner by a sudden irruption of the river in which case the title of the original owner continues unless abandoned, or whether it applies also to cases where the original owners' title is lost by the gradual slow and imperceptible process of diluvion, we need not pause to enquire, as it does not appear on the evidence in this case that the encroachment was by such slow and imperceptible degrees as to destroy the respondents' title to the land submerged. The learned Government Pleader, however, argued that while the principle of re-formation in situ might be applicable to the extent to which the respondents' portion of the old Balaramuni lanka or the Merka lanka as shown in the survey map of 1863 could be traced into the island now in dispute and identified with any part of it, the principle could not be invoked in respect of the increments to these lankas shown in the plan of 1878. It is difficult to appreciate this distinction. When once these accretions are held to belong to the respondents, as I think they should be for the reasons already indicated, no logical differentiation can be made between the respondents' title to the original lankas and their title to the subsequent accretions thereto. Their title to both must, in quality and scope, stand on the same footing, and there would seem to be no reason why the respondents should be denied the benefit of the re-formation of the accreted land any more than the re-formation of the original nucleus. The learned Government Pleader has also urged that the principle is not applicable unless the new formation has clearly ceased to be part of the river-bed and has emerged out of the surface of the stream in a cultivable or usable state. It may be mentioned here that while the central portion of the island in dispute, is fit for cultivation and is actually cultivated, being covered by alluvial deposits, there is a wide fringe of sandy waste all around. It was said that these sand shoals are still part of the river-bed and that the respondents could not claim title to any portion thereof, even if such portion could be identified with the site of the old lankas washed away. Reference was made to the definition of river-bed in Hindson v. Ashby (1896) 1 CH. 78, where, dealing with a claim to a strip of land alleged to be an accretion to the land of a riparian owner, Romer, ]., observed that the question whether any part of river-bed had lost its character as such and has accreted to dry riparian land was one of fact to be determined not by any hard and fast rule, but by regarding all the material circumstances of the case including the fluctuations of the river, the nature of the land, and its growths and' user, an observation which was approved by the Court of Appeal in Hindson v. Ashby (1935) 69 M.L.J. 171. In this view, it was said, the surrounding sand shoals could not, by. their very character and appearance, be said to have ceased to be part of the bed of the river in the ownership of the Crown. This may well be a material consideration if these portions were claimed as accretions; for accretion implies that a portion of the river-bed becomes the property of the adjoining owner by a process of gradual and imperceptible recession of the stream, and it is but common justice in such a case that the land claimed to have thus accreted should be shown to have definitely ceased to be part of the bed before the title of the owner of the bed could be Said to have been lost. But the doctrine of re-formation in situ rests on the principle hat the title of the original owner to the land washed away or submerged is not absolutely extinguished unless he has abandoned it but continues to remain in such owner, provided such land can be identified by boundaries, landmarks or otherwise, subject of course, to such easements of navigation or other rights to which the flow of water thereon may give rise. When the water has receded and the land re-appears again the original owner is held entitled to take the property if it can be clearly identified. This being the reason of the rule, it is difficult to see why it should be regarded as necessary that the land encroached upon by the river should be shown to have ceased to be part of the river-bed before it can be claimed by the original owner as his property. There is no question of recovering, in such cases, what did not originally belong to the owner. I am, therefore, of opinion that the respondents are entitled to claim such portion or portions of the island in dispute as can be proved to have formed on the ascertained site of their old lankas or of the accretions thereto shown in the survey plans of 1878.
31. The respondents' claim to the other block of land called the Bellampudi lanka is the subject of the connected appeal. The respondents' ownership of the bed to the middle of the stream has already been negatived. The only other ground of claim disclosed in respondents' pleading is an alleged re-formation on the site of an old lanka which originally belonged to them. There are no satisfactory materials to establish the identity of the new formation with the site of any old submerged lanka which once belonged to respondents. An attempt was, however, made before us to show that the land in dispute was an accretion to their riparian land. But no such case was put forward in the plaint, the reference to lateral accretion in paragraph 6 being obviously to 'any excess in area' in the new formation. In cases of this kind it is essential that parties should be strictly confined to the case which they make in their pleadings, for claims based on accretion, on re-formation in situ, or on riparian proprietorship up to the middle of the stream, involve different considerations and require different media of proof. In Sree Eckowrie Sing v. Heeralall Seal (1868) 12 M.I.A. 136, their Lordships of the Privy Council observed:
The plaintiffs must succeed or fail on their title to the land as alluvial. It is not competent for them now, the cause having been decided on this title to raise at the hearing of their appeal a different case, vis., one simply of original ownership of the site of the lands re-formed. Had that been the case alleged, some defence might have been made, founded on the nature of a boundary river, the ownership of its soil, the character, sudden or gradual of the original loss of land, and the effect of change from such causes in the land itself on the ownership in the soil.
32. In the present case, the respondents based their claim to the land as a re-formation of the site originally owned by them. It is not therefore competent for them now to raise a different case, namely, that the new formation is an alluvial accretion to their riparian land. Nor have they substantiated this new ground of claim.
33. I concur in the order made by my learned brother.