This was one of several real actions, depending on the same facts and principles ; and the parties to the others have agreed to abide by the decision in this.

The case comes before the court upon a report of the chief justice, before whom the cause was brought to trial at Northampton.. The evidence being given in, by agreement of parties, the case was taken from the jury without a verdict, to be submitted to the whole court upon the evidence to be reported. It was further agreed, that upon the grounds to be stated by the court, one or more commissioners should be appointed, to make surveys, draw a plan, and make a division of the land, conformably to the principles announced by the court, and under such direction as they might give, and report thereon to the court; and when the report should' be accepted, judgment in this and the other cases to be rendered accordingly.

The land in controversy is newly-made land, formed in what was formerly the bed of Connecticut Eiver, lying between the towns of Hatfield and Hadley. It has been gradually formed, in consequence of a change in the bed of the river, in the manner hereafter stated. The demandants, trustees of an incorporated academy, are owners of a tract of land in Hadley, on the east side of Connecticut Eiver, known as the school meadow land, bounded formerly by a curved line projecting considerably into the river. As long ago as 1805 or 1806, the water, in high freshets, began to find its way across the school meadow land. This increased from year to year, until the current was formed that way; and in 1825 a great portion, if not the main body of the stream, passed that way, thus taking a more direct line across, instead of following the former bend of the river. This continued to increase until it became the main channel, and the current through the old passage ceased. The new channel, thus formed, cut off and insulated the most projecting part of the school meadow land; the part, thus left, remained unchanged in position, and became an island, forming the right bank of the new stream as far as it extended.

In the mean time land began to form in various places in the old bed of the river thus deserted by the current, between *546the school meadow lands, thus insulated, on the east side, and the Hatfield shore on the west. The old channel, or deepest part of the rive»’ ^efore the change, was not in the middle of the river, buc nearest the Hatfield shore.

There was some conflicting evidence as to the place at which the alluvial land began to form; it being contended by the demandants that it began to form on the upper part of the island, and extended over towards and near the Hatfield shore. On the contrary, it was contended by the tenant that it began to' form near the Hatfield shore, and extended thence into the river. But this was considered immaterial, the chief justice being of opinioh and proposing to instruct the jury, that the demandants, as riparian proprietors, could not claim any of the land in controversy, being the land newly formed between the island and Hatfield shore, more northerly, that is higher up river, than their land extended before the formation of the land in controversy, and if it had so extended since the change in the course of the river, that fact might be inquired into by the commissioners to be appointed.

Several rulings and proposed instructions to the jury are specified in the report, which it is not necessary here to state, because they were substantially adopted, and are restated in the opinion of the full court.

The tendency of the evidence was, to show that in all that part of the river bed, which had become stagnant by the. change of current, sediment began to settle and accumulate, shoals and islets arose, detached, at first, but gradually uniting with each other; when above water, vegetation came in, and ultimately land was formed, which was available for use and valuable.

The question is, whose property are the lands thus newiy formed, and upon what legal principles shall the right of property in them be ascertained ?

It appears to be well settled by the law, both of this country and of England, and founded on the rule of the civil law, that on rivers not navigable, the right of the proprietors of the land on each side extends to the thread of the river, or middle *547line of the stream. This is taken to be the law in this commonwealth, subject to certain rights of the public to use the water for boating and rafting, and subject also to regulations in regard to fisheries. These modifications or exceptions are not applicable to the present question.

This question arises respecting the right of the riparian proprietors to certain lands formed in the bed of Connecticut River, between Hadley and Hatfield. This land, it appears by the case, was formed on a part of the soil which formerly constituted the deep bed and channel of the river, where the main current of water formerly flowed, in consequence of the river having changed its course and taken a new channel on the Hadley side. The effect of this has been, that the main body of the water has for some years flowed in a new channel, by means of which the water on the old bed of the river became stagnant, deposits of earth and sand were formed in various parts of it, which have gradually risen above the surface and united with each other so as to become valuable land. The question is, whose is this land ?

It has been repeatedly settled, both in this state and in Connecticut, that the Connecticut River, though valuable for the purposes of boating and rafting, yet, so far as riparian proprietorship is concerned, is considered a river not navigable, as that term is used in the common law. Adams v. Pease, 2 Conn. 481; Bardwell v. Ames, 22 Pick. 333.

The general rule, as a rule of the common law of England, was long since laid down as unquestionable by Lord Holt, who says, in the case of Rex v. Wharton, Holt, 499, that a river, of common right, belongs to the proprietors of the land between which it runs, to each that part nearest his land. This has been frequently, if not uniformly, adopted as the established rule. Bac. Ab. tit. Prerogative; Sir John Davies’s R. 155. And the same rule has been repeatedly declared and adjudged in this commonwealth. It is derived mainly from the rule, that the riparian proprietor is owner of the soil under the water, and by the general law of property becomes entitled as of right to all accessions.

The general rule is recognized and established in thi, *548commonwealth, in the leading case of Ingraham v. Wilkinson, 4 Pick. 268. It is a case which goes far to settle principles which must govern the present. It recognizes the rule of the common law, that the property in the soil of rivers not navigable, subject to public easements, belongs to those whose lands border upon them; and from this right of property in the soil in the bed of the river the court deduce the right of property in an island, which gradually arises above the surface and becomes valuable for use as land. Assuming the thread of the river, as it was immediately before such island made its appearance, this rule assigns the whole island, or bare ground formed in the bed of the river, if it be wholly on one side of the thread of the river, to the owner on that side ; but if it be so situated that it is partly on one side and partly on the other of the thread of the river, it shall be divided by such line, and held in severalty by the adjacent proprietors.

This seems to be clear and intelligible, and has a direct bearing upon the present cáse. But the difficulty in applying the rule arises from this consideration, that, from the very nature of things, the thread of the river may itself be frequently changed by circumstances. "Where there is a gradual accretion on one shore, forming an alluvion and permanent addition to that shore, it must tend to change the thread of the river, by carrying the medium line towards the opposite shore. If, at the same time, with such accretion on one side, there be a wearing away of the opposite bank, the thread of the stream will be moved an equal distance in that direction. But if the river does not wear away on the opposite side, then the whole bed is narrowed, and the thread of the river will be moved to an extent equal to one half of the new formation, and in the same direction.

So in the case just now supposed, of an island arising in the middle of the river, it is divided by that line which was the thread of the river immediately before the rise of the island. But that line must thenceforth cease to be the thread of the river, or filum aqum, because the space it occupied has ceased to be covered with water. But, by the fact of an island being formed in the middle of the river, two streams *549are necessarily formed by the original river dividing it into two branches; the island itself, having become solid land, forms itself a bank of the new stream on the one side, and the old bank on the main shore forms the other. And the same rule applies on the other side of the island. There must, then, be a filum aquae to each of these streams, whilst the old filum aquae is obliterated to the extent to which land has taken the place of water. But this island, having all the characteristics of land, may soon be divided and subdivided, by conveyances and descents, and all the modes of transmission of property known to the law, and thus become the property of different owners. Now suppose another island formed in one of these branches, between the first island and the original main shore. It seems to us that it must be divided upon the same principle as the first; but, in doing it, it will be necessary to assume as the filum aquae the middle line between the first island and the original river bank on that side.

If this is a correct view of the practical consequences flowing from the adoption of the principle stated, and it appears to us that it is, an obvious difficulty presents itself, in making that line a fixed standard for the demarcation of the boundaries of real estate between coterminous proprietors, which is itself fluctuating and changeable. Perhaps a satisfactory answer to this may be found in the suggestion, that the rule is equitable, and as certain as the proverbially variable nature of the subject-matter will admit; and, in adapting it to the varying circumstances of different cases, a steady regard must be had to the great principle of equity, that of equality.

This changing of the filum aquae seems not to be distinctly treated in any case, but it seems that it must necessarily occur In many cases. In addition to those already mentioned, suppose a river, by slow accretions or washing away, widens or narrows on both sides as it may, but unequally, the filum aquae must change its actual line. Suppose an island dividing a river for some distance shall be wholly washed away, the filum aquae must shift and pass along a line which was formerly solid land. In a passage cited by Chief Justice Parker, in Ingraham v. Wilkinson, from lord Hale, such a shifting of *550the filum, aqua in one ease is alluded to. The passage is this : “ If the filum aqua divide itself, and one part take the east and the other the west, and leave air ’«land in the middle between both fila¡ the one half will belong to the one land and the other to the other.” And Lord Hale adds further, in the same connection, “ that this is to be understood of islands newly made; for if a part of an arm of the sea — and the same thing is true of a river, which is material to the present case — by a new recess from his ancient channel, encompass the land of another man, his propriety continues unaltered.” Hargrave’s Law Tracts, 37.

It may be added here, on the authority of-lord Hale, that he derives the title to islands, in creeks or havens or arms of the sea, from the right of property in the soil under the water, stating that this is primá facie and of common right in the king; yet if, in point of propriety, it doth belong to a subject, by grant or prescription, the islands that happen within the precincts of such private propriety of a subject, will belong to the subject. This is applicable, by strict analogy, to the case of a river not navigable, when the right of property is admitted to be in the riparian proprietor, ad filum aqua.

If this is a correct view of the general principle of law, and its modifications under certain changes, it seems to us sufficient to settle the present case. It appears that the demand-ants have long been owners of a tract of land situated in Hadley, and forming, for some distance, the east bank of Connecticut River. The tenant and other proprietors were owners of lands in Hatfield, forming the west bank or shore of Connecticut River, opposite the school meadow land. Many years ago the water of the river began to find its way, at first in small quantities and in high freshets, over the school lands, till at length it formed a deep cut, and ultimately the main or sole channel of the river. A small portion of the school lands, forming the original east bank of the river, was not washed away, but remained unaffected by the change. The old channel, on the west side of this portion of the school land, now become an island, being deserted by the current, began very slowly to fill up, and after many years, say twenty *551or more, became covered with vegetation; and this is the land which is the subject of this controversy.

Now, as to this island, it is not newly made, but a portion of the old solid school land ; it remains unchanged in site or character, and is, as it was before, the property of the demand-ants. It also remains a fixed portion of the easterly bank of that branch of Connecticut River, which now lies between that island and the Hatfield shore. There is nothing to show that the filum aqua has changed to the extent up and down the river, to which this island still extends. Each who was then riparian proprietor owned the soil under the river to that filum aqua or thread of the river, and the newly-formed land will belong to them respectively to that line.

In regard to the newly-formed land, if any, which lies further north, or further up stream than the head of the island constituting a part of the old school land, there is certainly more difficulty.

If before these sand-banks, flats and islands began to form in the old bed of the river, this island had extended upwards, by alluvion strictly so called, it might have been deemed part of the school lot and of the island, and would have modified and regulated the filum aqua to a like extent upwards. By alluvion here, we mean such slow, gradual, and insensible accretion, that it cannot be shown at what time it occurred ; but we are not aware that it is claimed that this insular portion of the demandants’ land was thus extended upwards by such insensible accretion, previously to the formation of the land in question. If it be claimed that it was, it will be a proper subject to be inquired into and reported on by the commissioners. When the head of the island, or school meadow land is ascertained, a line is to be struck across the river, at right angles to its general course, at such head of the island, and in all that part of the stream below such line, the thread of the river will be the middle line between the island and the Hatfield shore.

Then we must consider that part of the river lying above the island; and here the filum aqua was probably modified by the great change in the bed of the river, in effect forming two *552streams, one each side of the insulated part. It is manifest that, by breaking out a new channel to the east, and yet retaining, at least for a considerable time, its old channel on the west, the river above the dividing point for some distance was more or less widened. It was a very slow process by which, in the more sluggish current on the west side, shoals, sand-bars, islets, and ultimately firm land were formed. In the mean time, the river had formed for itself a new east bank on the Pladley side, the old west bank remaining as it was on the Hatfield side. So long as that state of things continued, the filum aqua would naturally be extended, forming the middle line between such old west bank and such new east bank. If this east bank was thus formed, and the thread of the river above the island thus was extended before the new land formed, then it appears to us that the law of proprietary division must be determined by the filum aquas, as thus extended easterly, and that this rule must be applied to all that part of the island above a line drawn at right angles with the river, across it at the head of the island. All lying westerly, or on the Hatfield side of the filum aquae thus ascertained, if any, belonging to the Hatfield proprietors, and all easterly of such filum aquae to the school meadow proprietors, or the proprietors lying on that side.

In ascertaining the thread of the river, it will be proper to take the middle line between the shores upon each side, without-regard to the channel, or lowest and deepest part of the stream. And in ascertaining the shores, or water lines on each side, to measure, it will be proper to find what those lines are when the water is in its natural and ordinary stage at a medium height, neither swollen by freshets or shrunk by drought.

These views, we think, embrace all the legal principles necessary to determine the right of property in these newly-formed lands, as between the opposite riparian proprietors. The division of these lands amongst the Hatfield proprietors themselves, will be regulated by the rule laid down in the case of Deerfield v. Arms, 17 Pick. 41. The effect of that rule is, to give each one a length on the new water line proportioned *553to his length on the old water line, whether the one be longer or shorter than the other. In applying that rule in this case, each Hatfield proprietor will have a line on the thread of the river, when ascertained, proportioned to his line on the Hatfield shore before the river filled up.

Pursuant to the agreement of the parties, one or more commissioners are to be appointed, unless agreed on by the parties, to survey the land, lay down lines for the division of the same, agreeably to the principles thus stated, and malr-’ their report to the court.