Opinion

In response to the mounting social, environmental and health crises of recent years, governmental authorities have considerably expanded the planning and construction of “public improvements.” Because the definite commencement of a public project is almost invariably preceded by significant publicity and public interest, land values in the vicinity of the potential project often will increase in response to this foreknowledge. A recurring issue in eminent domain litigation is whether, and to what extent, such increases in land values attributable to the proposed project comprise a proper element of the “just compensation” to be paid to a land owner if his land is ultimately taken for a project. This question has not been definitely resolved by California decisions to date; three cases before our court today require us to confront this issue of the proper interpretation of our constitutional “just compensation” clause directly, and additionally require us to probe the practical problems of application attending our constitutional conclusions.

For the reasons discussed hereafter, we have concluded that the few appellate decisions which have intimated that any increase in value arising from the expectation of the coming project should be excluded from just compensation must be reexamined in light of the realities of a landowner’s position. In the early stages of a desirable project’s development, land which is expected to be within the vicinity of the project, but is not expected to be taken for the project, will naturally increase in value, and a landowner who chooses to sell such land at this time will gain the benefit of this incremental value; similarly, one who buys such land at this time must pay this incremental amount for his purchase. It is not until a particular piece of property is reasonably expected to be condemned for the project that this enhanced market value, attributable to the land’s anticipated proximity to the improvement, disappears. We have determined that it would be unfair, in computing just compensation, to eliminate the appreciation in market value which a specific piece of property in fact enjoyed before it was designated for condemnation, since that would in effect deny to the owner the market value of his property prior to the time it was pinpointed for taking.

1. The facts of the instant case.

Mrs. Mazie Woolstenhulme, defendant-landowner in the instant eminent domain action, owns a ranch of approximately 13,150 acres in a remote portion of Mariposa County. One end of the ranch borders Lake McClure, an artificial lake created in 1927 and owned by Merced Irrigation District, the condemner in this proceeding. In the present action, the district condemned 189 acres of defendant’s land for use in connection with a new, multipurpose water project planned for the region. The jury awarded defendant $250 per acre for this land, and the district attacks this valuation on appeal.

Prior to the commencement of the district’s new water project, little domestic water and no power was available in the Lake McClure region; land in the area was largely uninhabited and devoted primarily to cattle grazing. Lake McClure was subject to wide seasonal fluctuation, covering a maximum of 2,700 acres during the winter months, but contracting to merely 30 acres, surrounded by mudflats, in summer. The district owned a buffer strip of 200 feet around the lake, presumably adjacent to the lake’s border in its high water stage. Evidence introduced at trial revealed that, during this pre-improvement stage, land in the area had not sold for higher than $125 an acre.

In the late 1950’s the district began evolving plans for a new Lake McClure project that was considerably to alter the nature of the area. The new project was to increase the size of the lake, and eliminate most of the fluctuation in its coverage and depth; it was to provide the neighboring lands with power and domestic water not available from the old dam and lake. By 1962 the district had begun a quest for federal funds to assist in the financing of the project, and early in 1963 several newspaper articles informed the public that the completed Lake McClure project would include recreational facilities, such as camping, boating and fishing. The trial court found that about January 1, 1963 the public, while unaware of “exactly what area, what spots were to be recreation,” did know of the general recreation plans, and that, as a result, property values in the area began to increase within a short time thereafter. The court also found that by January 1, 1965 the plans for the project had progressed to a point where it became “reasonably probable” that the present parcel of defendant’s land would be taken for the project. During 1965 and 1966, a flurry of land sales occurred in the area at prices ranging from $250 to $600 an acre. The district filed the amended complaint on which this action is based in August 1967.

At trial plaintiff condemner’s appraisal witness testified that, omitting consideration of the new Lake McClure project, cattle grazing was the highest and best use of the 189 acres in question, and he valued the land, on the basis of the normal market value of such land in the past, at $125 an acre. Mrs. Woolstenhulme, the defendant-landowner, stated that in her opinion the property had a value of $600 an acre; she admitted, however, that in February 1966 she had sold a similar parcel of her ranch for $250 an acre. Defendant’s expert appraisal witness, Richard Leuschner, testified that when used for grazing purposes as part of defendant’s ranch, the land would have a value of $200 an acre. Leuschner declared, however, that viewing the 189 acres as a separate tract, “development,” rather than cattle grazing, was the highest and best use of the property and he stated that, on the basis of his examination of sales of comparable properties, he would evaluate defendant’s land at $600 an acre, after deducting $50 an acre of “enhanced value” arising from the Lake McClure project.

In attempting to explain this surprisingly small increment of value which he attributed to the pending improvement, Leuschner testified that he believed that the new Lake McClure project was only one of a considerable number of factors resulting in the rapid increase in land value in the region, and was not an overwhelming factor at that. The appraiser described a growing statewide trend, stretching over almost a decade, of sales of agricultural foothill property to city residents seeking a country “home away from home”; he attributed the trend, in large part, to the tremendous population increase in California’s urban centers in recent years. Leuschner also testified that although Mariposa County is relatively far removed from the heavily populated areas of Los Angeles and the Bay region, newly constructed freeways had reduced the traveling time considerably and had made the region accessible for “recreational development” purposes. The appraiser concluded that even without the new water project, the area would have been an attractive “development” site, for he considered the old lake adequate for swimming and fishing.

In support of Leuschner’s valuation, defendants offered evidence of some of the 1965 and 1966 sales of neighboring parcels as “comparable sales” under section 816 of the Evidence Code. The district objected to the introduction of these sales on the grounds that the sale prices reflected an increase or enhancement in value attributable to benefits created by the very project for which condemnation was sought, an enhancement which the district contended was not a proper element of “just compensation.” The condemner strongly disputed Leuschner’s analysis of the increase in land values in the area, and argued that it was the new project which had transformed land, previously useful only for grazing, into valuable lakefront sites. The trial judge, although finding that the proffered sales reflected “substantial enhancement” due to the recreational potential of the project, nevertheless admitted the evidence, indicating that he would instruct the jury to eliminate any post-January 1, 1965 enhancement attributable to the project from the determination of just compensation. The jury was so instructed, and, as stated above, awarded defendant $250 an acre.

On this appeal the district raises two principal objections to the trial court’s valuation rulings. First, the district contends that the court erred in instructing the jury to exclude only that “enhancement value” which arose after January 1, 1965. The district asserts that the general rule in this state is that, in determining just compensation, all “enhanced value” attributable to the condemner’s proposed improvement must be excluded and that the court erred in permitting defendant to recover the pre-1965 increment in value which resulted from public knowledge and expectation of the Lake McClure project. Second, the district contends that, even assuming that pre-1965 enhancement was a proper element of compensation, the trial judge erred in admitting evidence of sales which were found to reflect “substantial” post-January 1, 1965 enhancement. Plaintiff asserts that such sales are not “comparable sales” within the meaning of section 816 of the Evidence Code, and thus are inadmissible.

As explained below, we have concluded that neither of plaintiff’s objections should be sustained. We shall initially point out that, under our just compensation clause, an owner of the condemned property should be compensated for the increase in value which his land has experienced in anticipation of the benefits of a proposed improvement, so long as it is not reasonably probable that the specific piece of property being evaluated is to be taken for the improvement. Secondly, we shall explain that under Evidence Code section 816, sales are not necessarily “non-comparable” simply because they reflect “substantial" project enhancement, and thus a trial court, in exercising the discretion granted by the statutory provision, may properly admit such sales in evidence.

We turn first to the proper measure of just compensation in these circumstances.

2. The trial court did not err in permitting the jury, in determining just compensation, to consider the “project enhanced” value which accrued to defendant’s property prior to the time that it was reasonably probable that the property would be taken for the improvement.

(a) A legitimate element of just compensation lies in the increase in value resulting from a reasonable expectation that a particular piece of property will be outside a proposed public improvement, and thus will reap the benefits of that improvement.

Article I, section 14 of the California Constitution provides that “Private property shall not be taken or damaged for public use without just compensation having first been made to . . . the owner . . and although the constitutional provision does not explicitly define the measure of “just compensation,” it has long been established that in general “the compensation required is to be measured by the market value of the property . . at the time of the taking. (Rose v. State of California (1942) 19 Cal.2d 713, 737 [ 123 P.2d 505 ]; see, e.g., Muller v. Southern Pacific Branch Ry. Co. (1890) 83 Cal. 240, 243, 245 [ 23 P. 265 ]; Spring Valley Water Works v. Drinkhouse (1891) 92 Cal. 528, 533 [ 28 P. 681 ]. See also Code Civ. Proc., § 1249.) “Market value,” in turn, has been defined as “the highest price estimated in terms of money which the land would bring if exposed for sale in the open market, with reasonable time allowed in which to find a purchaser, buying with knowledge of all of the uses and purposes to which it was adapted and for which it was capable.” (Sacramento So. R.R. Co. v. Heilbron (1909) 156 Cal. 408, 409 [ 104 P. 979 ].)

The “market value” of a given piece of property, of course, reflects a great variety of factors independent of the size, nature, or condition of the property itself. The general character of the neighborhood, the quality of the public and private services, and the availability of public facilities all play important roles in establishing market value. Thus, widespread knowledge of a proposed public improvement, planned for an indefinite location within a given region or neighborhood, will frequently cause the market value of land in the region or neighborhood to rise. Such an increase in market value results from the expectation that a given parcel of property will be outside of the project and will soon enjoy the benefits of the proposed improvement. If, for example, the planned project is a public park, land in the vicinity will be expected to gain the advantages of a nearby recreational area, and will consequently become more desirable and more valuable.

Sometimes, however, property which has increased in value, out of an initial anticipation that the land would be outside of a public improve ment, must itself be taken for the construction or creation of that public improvement. Since the instant case presents that situation, our first issue must be to determine whether, in such a case, the owner of the land to be taken should be compensated for the loss of this increase in value—an increase that occurs prior to the time that it is known the particular piece of property will be included in the project.

We note at the outset that, although this court has not spoken directly to the issue in the past, the majority rule in other jurisdictions is that such “project enhanced” value does constitute a proper element of value for which the landowner is entitled to be compensated. (See 4 Nichols on Eminent Domain (3d ed. 1962) § 12.3151[2], pp. 209-210.) Most notably, the United States Supreme Court has consistently construed the “just compensation” clause of the Fifth Amendment of the federal Constitution to countenance the landowner’s recovery of this “project enhanced value” unless his property was itself “probably within the scope of the project from the time the Government was committed to it.” (United States v. Miller (1943) 317 U.S. 369, 377 [ 87 L.Ed. 336, 344 , 63 S.Ct. 276 , 147 A.L.R. 55 ]; see Kerr v. South Park Comrs. (1886) 117 U.S. 379, 384-386 [ 29 L.Ed. 924, 926-927 , 6 S.Ct. 801 ]; Shoemaker v. United States (1893) 147 U.S. 282, 303-305 [ 37 L.Ed. 170, 186-187 , 13 S.Ct. 361 ]; United States v. Reynolds (1970) 397 U.S. 14, 16-18 [ 25 L.Ed.2d 12, 15-17 , 90 S.Ct. 803 ].) The courts of our sister states have generally embraced a like position. (See, e.g., Williams v. City & County of Denver (1961) 147 Colo. 195, 200 [ 363 P.2d 171, 174 ]; Cole v. Boston Edison Company (1959) 338 Mass. 661, 666 [ 157 N.E.2d 209, 212 ]; Andrews v. State of New York (1961) 9 N.Y.2d 606 [ 217 N.Y.S.2d 9 , 176 N.E.2d 42 ]; Rowan v. Commonwealth (1918) 261 Pa. 88, 94-95 [ 104 A. 502, 504-505 ]; Stafford v. City of Providence (1873) 10 R.I. 567, 571-572 ; State v. Wood (1969) 22 Utah 2d 317, 318-320 [ 452 P.2d 872, 873-874 ].)

In our view, the widespread agreement on this point finds firm support in the principle that “market value” is the proper measure of just compensation, and, for the reasons explained more fully below, we now join these sister states in holding that this kind of “enhancement value” is a proper element of just compensation.

On this appeal the district, although not contesting the general validity of the market value standard of “just compensation,” contends that California precedent has long established “that in arriving at a determination of . . . market value ... it is not proper to consider the increase, if any, in the value of such land by reason of the proposed improvement which is to be made on the land by the condemner.” (County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 78 [ 291 P.2d 98 ].) The district claims that this doctrine, derived from a statement by this court in San Diego Land etc. Co. v. Neale (1888) 78 Cal. 63, 74-75 [ 20 P. 372 ], precludes a jury from including in an eminent domain award any increase in value “attributable to” the proposed project (or, as it is often referred to, “project enhanced value”). In support of its position the condemner relies on a series of Court of Appeal decisions, which contain dicta to- the effect that “[a]ny rise in value before the taking . . . caused by the expectation of that event” is to be disallowed in computing just compensation. (City of Pasadena v. Union Trust Co. (1934) 138 Cal.App. 21, 26 [ 31 P.2d 463 ]; People ex rel. Dept. Pub. Wks. v. Shasta Pipe etc. Co. (1968) 264 Cal.App.2d 520, 539 [ 70 Cal.Rptr. 618 ]; People ex rel. Dept. Water Resources v. Brown (1967) 255 Cal.App.2d 597, 599 [ 63 Cal.Rptr. 363 ]; Community Redevelopment Agency v. Henderson (1967) 251 Cal.App.2d 336, 343 [ 59 Cal.Rptr. 311 ]; County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 78 [ 291 P.2d 98 ].) Under this line of cases, the condemner argues, the general increase in neighborhood land values which, frequently accompanies the announcement of a desirable public improvement constitutes “project enhanced value” for which the landowner is never entitled to be compensated; in sum, the benefit conferred upon the land by the condemner should not be charged against the benefactor.

This position, based on an expansive interpretation of the concept of “project enhanced value,” which past decisions have indicated is to be excluded from compensation, obscures pertinent distinctions between different types of “project enhanced value.” The value of land can be said to increase “by reason of the proposed improvement” (County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 78 [ 291 P.2d 98 ]) for at least three distinct reasons: (1) the worth of property known to be within the project may rise when the land is valued as part of the proposed improvement rather than as a separate tract of land; (2) the value of property expected to be condemned may rise because of the anticipation that the condemner will be required to pay an inflated price for the land at the time of condemnation; and (3) the value of property expected to be outside of the proposed improvement may rise because it is anticipated that the land will reap the benefits resulting from proximity to the coming project. Although past California decisions- have not found it necessary to distinguish between these various “increases in value,” the district’s contention in the instant case brings the need for such analysis into sharp focus. We shall analyze each of these three situations in the course of this opinion.

We begin with the seminal decision of San Diego Land etc. Co. v. Neale (1888) 78 Cal. 63 [ 20 P. 372 ]. In Neale , defendant’s land was being condemned as a reservoir site in connection with the construction of a dam on a neighboring tract. At trial, the condemnee asked his appraiser to evaluate the land on the basis of its use as a reservoir site, taking into account the on-going construction of the dam. In holding this question improper on appeal, the Neale court declared: “it seems monstrous to say that the benefit arising from the proposed improvement is to be taken into consideration as an element of the value of the land. . . .” In context, this statement, which gave rise to the doctrine relied on by the district in the instant case, clearly is no more than a declaration of the firmly established premise that “compensation is based on loss imposed on the owner, rather than on benefit received by the taker. [Citations.] The beneficial purpose to be derived by the condemner’s use of the property is not to be taken into consideration in determining market values, for it is wholly irrelevant.” (People v. La Macchia (1953) 41 Cal.2d 738, 754 [ 264 P.2d 15 ]; see City of Stockton v. Vote (1926) 76 Cal.App. 369, 404 [ 244 P. 609 ]; Boston Chamber of Commerce v. City of Boston (1910) 217 U.S. 189, 195 [ 54 L.Ed. 725, 727 , 30 S.Ct. 459 ].) Thus, the improper “enhancement” or “benefit” referred to in Neale is simply the increase in value which a condemned tract gains when it is valued as part of the proposed project, i.e., the first type of “project enhanced value” referred to in the preceding paragraph. It is clear, of course, that this incremental value is one which could never be considered in determining “just compensation” under the established definition of “market value” set out above.

We turn to the second aspect of “project enhanced value” which we have noted in the trilogy outlined supra. A situation in which the enhanced value of the land should not be included as compensation occurs when the increased value is due to speculation based upon the imminence of a taking. After a parcel of land has been designated for condemnation, the “actual market value” of the parcel will frequently fluctuate as a result of the impending condemnation. An increase in the value of property which can reasonably be expected to be condemned can generally be explained only as a result of speculation by potential purchasers that the condemner may be compelled to pay an artificially inflated price for the property. (See Palmer, Manual of Condemnation Law (1961) § 154.) Although this speculation does, in a sense, affect “actual market value” (see 1 Orgel on Valuation Under Eminent Domain (2d ed. 1953) § 83, p. 355 et seq.), this is not the “open market” value contemplated by our controlling decisions (e.g., Sacramento So. R.R. Co. v. Heilbron (1909) 156 Cal. 408, 409 [ 104 P. 979 ]; cf. United States v. Cors (1949) 337 U.S. 325 , 333 [ 93 L.Ed. 1392 , 1399, 69 S.Ct. 1086 ]). Almost all courts universally agree that such an increase in value, based on a purchaser’s conjecture of what the condemner may ultimately be required to pay, is not a proper element of “fair market value” for “just compensation” purposes. (See, e.g., United States v. Reynolds (1970) 397 U.S. 14, 16 [ 25 L.Ed.2d 12, 15 , 90 S.Ct. 803 ]; United States v. Miller (1943) 317 U.S. 369, 377 [ 87 L.Ed. 336, 344 , 63 S.Ct. 276 , 147 A.L.R. 55 ]; Olson v. United States (1934) 292 U.S. 246, 261 [ 78 L.Ed. 1236, 1247 , 54 S.Ct. 704 ].) If a tribunal were required, in setting just compensation, to consider an increase in value arising merely from the anticipation of the tribunal’s final award, then logically a speculator would in effect be able to set “just compensation” through his own purchase price. (See 1 Orgel on Valuation Under Eminent Domain (2d ed. 1953) § 83, p. 359.) In our view this type of “enhanced” value is clearly not a legitimate element of just compensation and thus we now reiterate that such increases in value cannot properly be taken into consideration in determining the fair market value contemplated by our constitutional just compensation requirement.

The (1) “enhanced value” arising from the condemner’s potential use of the property itself for the project, as in Neale , and (2) the “enhanced value” resulting from speculation over the amount of an imminent condemnation award are clearly distinguishable, however, from (3) the increase in land values of property which is expected to be adjacent to or near a proposed project. This category is the third in the grouping set out above. Although the increase in value of the adjacent or nearby property is undoubtedly “attributable” to the project, it results not from the expectation that the land will be taken for the project, as in the case of the property in Neale , which is included in the project, or of the property which enjoys the speculative gain, but instead from the expectation that the land will not be taken for the project. It is this distinction which the argument of the condemner in the instant case ignores, and upon which, we have concluded, plaintiff’s position founders.

The difference between the project enhanced value of the adjacent property and that of the other two situations discussed above is that the rise in value of the adjacent property is a legitimate element of its “fair open market value.” Clearly, the expected proximity of a tract of land to a proposed project constitutes a factor “which a buyer would take into consideration in arriving at a fair market value, were he contemplating a purchase of the property” (People ex rel. Dept, of Public Works v. Donovan (1962) 57 Cal.2d 346, 352 [ 19 Cal.Rptr. 473 , 369 P.2d 1 ]), and as such we think the value attributable to this anticipated proximity constitutes a proper element of just compensation. “The rule is, that the owner is entitled to the market value of his land, to be determined in view of all the facts which would naturally affect its value in the minds of purchasers generally. . . . ‘Any existing facts which enter into the value of the land in the public and general estimation, and tending [sic] to influence the minds of sellers and buyers, may be considered.’ [citation].” (Spring Valley Water Works v. Drinkhouse (1891) 92 Cal. 528, 533 [ 28 P. 681 ]; see Joint Highway Dist. No. 9 v. Ocean Shore R.R. Co. (1933) 128 Cal.App. 743, 753-759 [ 18 P.2d 413 ]; City of Stockton v. Vote (1926) 76 Cal.App. 369, 401-407 [ 244 P. 609 ].)

The courts have long held that benefits of government activities, reflected in market value, compose part of just compensation for land. Thus, increases in the value of a condemnee’s land “attributable to” a wide variety of activities paid for by government, or instituted at the behest of government, are properly includable in computations of just compensation. (See, e.g., People ex rel. Dept, of Public Works v. Donovan (1962) 57 Cal. 2d 346, 352-354 [ 19 Cal.Rptr. 473 , 369 P.2d 1 ] (“reasonable probability of a zoning change” a factor to be considered); County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 78-79 [ 291 P.2d 98 ] (increase in value from neighboring city improvements includable in determining value of tract to be taken for county project); City of San Diego v. Boggeln (1958) 164 Cal.App.2d 1, 6-7 [ 330 P.2d 74 ] (same).) Under these precedents the increase in value of lands expected to be outside a project constitutes a proper element of “just compensation.”

The district argues, however, that even if this increased value in neighborhood property is a valid component of “market value,” it should not be considered in determining “just compensation." Just compensation, the condemner asserts, is only intended to put the landowner in the same position he would have held if the project had not been built; the inclusion of this “enhancement” element in compensable value transgresses the principle that “just compensation” requires that compensation be “just” to the public as well as to the condemnee. (See People ex rel. Dept, of Public Works v. Pera (1961) 190 Cal.App.2d 497, 499 [ 12 Cal.Rptr. 129 ].) To require a condemner to pay for value which has arisen only because of its initiation of a project, plaintiff suggests, is to give the landowner a “windfall” at the expense of the public fisc.

We believe that the condemner’s argument rests upon its assertion that the basic purpose of “just compensation” is simply to return a landowner to the same position he would have held if the public project had never been constructed or contemplated. In positing such a purpose to our constitutional provision, however, the district has subtly assumed away the entire question at issue. Of course, as we have stated above, “just compensation” contemplates compensation measured by what the landowner has lost rather than by what the condemner has gained (People v. LaMacchia (1953) 41 Cal.2d 738, 754 [ 264 P.2d 15 ]). Nevertheless, the long-established recognition of “market value at the time of taking” as the general measure of “just compensation” reflects a deeply rooted judgment that, in determining just how much the landowner has lost, the state bears the responsibility of meeting the reasonable market evaluations of potential sellers or purchasers. General adherence to the “market value” measure insures a landowner that, in general, he will not be penalized for retaining his land after general public knowledge of the project. He should be assured that if his property is ultimately condemned, the condemner will compensate him for its “market value,” ideally at the price at which he could have sold the land on the open market just prior to the taking.

Inclusion of “project enhanced value” in compensation is essential if, in accordance with the above principle, the reasonable evaluations of landowners are to be met. In a situation in which the government decides, some time after the initial completion of a project, that expansion of the project is necessary, “just compensation” would clearly require that a condemnee, who had previously purchased his property at an increased price in the expectation that he would be near the improvement, should be compensated for “full” market value, including the increment paid for “project enhancement.” (See 4 Nichols on Eminent Domain (3d ed. 1962) § 12.3151 [3], pp. 210-211.) Since these owners purchased the property at the enhanced value, we could hardly justify the exclusion of this “enhanced” value from compensation if their property is ultimately taken.

For the same reason, the increase in value of land which is initially expected to be outside the boundaries of a proposed improvement, must be recognized to constitute a proper element of just compensation. Purchasers and sellers regularly, and quite reasonably, take into account the benefit that the land can be expected to> reap from an imminent public project, and it would be equally unfair and incompatible with the principles underlying our constitutional just compensation provision to exclude such enhanced value. Although the district chooses to characterize compensation for this project enhanced value as a “windfall” to the landowner, that epithet might equally be applied to the wide variety of other components of market value for which a landowner might not have directly “paid,” factors such as zoning laws, public services and general neighborhood appearance which, as previously noted, have long been recognized to be legitimate elements of “just compensation.”

In light of this analysis and the weight of authority, we now hold that increases in value, attributable to a project but reflecting a reasonable expectation that property will not be taken for the improvement, should properly be considered in determining “just compensation.”

The following Court of Appeal decisions are disapproved to the extent that they contain broad statements inconsistent with this conclusion: People ex rel. Dept. Pub. Wks. v. Shasta Pipe etc. Co. (1968) 264 Cal.App.2d 520, 539 [ 70 Cal.Rptr. 618 ]; People ex rel. Dept. Water Resources v. Brown (1967) 255 Cal.App.2d 597, 599 [ 63 Cal.Rptr. 363 ]; Community Redevelopment Agency v. Henderson (1967) 251 Cal.App.2d 336, 343 [ 59 Cal.Rptr. 311 ]; City of San Diego v. Boggeln (1958) 164 Cal.App.2d 1, 5 [ 330 P.2d 74 ]; County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 78 [ 291 P.2d 98 ]; City of Pasadena v. Union Trust Co. (1934) 138 Cal. App. 21, 26 [ 31 P.2d 463 ].

(b) The trial court properly instructed the jury to exclude all “project enhancement’ accruing after it was probable that the land to be valued would be taken for the project.

We have recognized above that under certain circumstances an increase in the value of land which is “attributable” to the proposed project may appropriately be included as just compensation. We also recognize that, in practice, the segregation of those cases in which “enhancement” should be compensable from those in which it should not will often entail a difficult task. To that problem we now turn.

In some instances the public may know from the time of the first announcement of the improvement that certain land will be included in the project. In such cases, since the public knows that the land will not receive the benefits of proximity to the project, the market value of the property will experience no such enhancement; thus, when such property is condemned, the landowner should not- receive any “project enhanced value.” “If it is known from the very beginning exactly where the improvement will be located if it is constructed at all, the property that will be required for its site will not participate in the rise or fall in values, for, since the property is bound to be taken if the improvement is constructed, it can never by any possibility either suffer from or enjoy the effects of the maintenance of the public work in its neighborhood; and consequently, it is well settled that in such a case in valuing the land the effect of the proposed improvement upon the neighborhood must be ignored.” (4 Nichols on Eminent Domain (3d ed. 1962) § 12.3151[1], pp. 205-206; see Note, Recovery for Enhancement and Blight in California (1969) 20 Hastings L.J. 622 , 629.)

Even when public information does not disclose definitely that a given piece of property will be used for the project, however, the landowner may not be properly entitled to “project enhanced” value. Governmental bureaucratic action is notoriously slow, and in many instances the public in general, and, in particular, interested landowners and potential buyers, will be able to determine accurately, well in advance of the formal acceptance of condemnation plans, that a given tract of property will probably be taken for the improvement. In such a case the market value of the land facing imminent condemnation will not rise because, as in the instance of “definite inclusion,” potential purchasers and sellers can reasonably foresee that the property will not enjoy the advantages of the coming improvement. As our earlier analysis demonstrates, the inclusion of “enhancement value” in compensation serves only to preserve the reasonable market value of the property. We see no reason to require the state to pay an incremental value if an informed individual could not reasonably expect that the property would be outside of the project. As the United States Supreme Court has stated in United States v. Miller (1943) 317 U.S. 369, 377 [87 L.Ed 336, 344, 63 S.Ct. 276 , 147 A.L.R. 55 ], enhancement value should not be includable in “just compensation” whenever the condemned lands “were probably within the scope of the project from the time the Government was committed to it.”

If, on the other hand, when plans for the proposed project first became public and when the consequent enhancement of land values began, the probability was that the land in question would not be taken for the public improvement, the landowner would be entitled to compensation for some “project enhancement.” During that period when it was not likely that his land would be condemned, the fair market value of the property may have appreciated because of anticipation that the land would partake in the advantages of the proposed project. The owner would be entitled to such increase in value. On the other hand, once it becomes reasonably foreseeable that the land is likely to be condemned for the improvement, “project enhancement,” for all practical purposes, ceases. Thus, in computing “just compensation” in such a case, a jury should only consider the increase in value attributable to the project up until the time when it became probable that the land would be needed for the improvement. (See United States v. 2,353.28 Acres of Land, etc., State of Fla. (5th Cir. 1969) 414 F.2d 965, 971 ; United States v. 172.80 Acres of Land, etc. (3d Cir. 1965) 350 F.2d 957, 959 .)

The approach prescribed by the trial judge in the instant case appears to accord with these standards. At the request of the parties, the trial judge conducted preliminary proceedings, prior to the empanelment of the jury, at which both parties presented evidence relating to the timetable of the Lake McClure project and to the inclusion of defendant’s land within that project. The trial judge concluded, first, that general public knowledge of the proposed recreational aspect of the project commenced in January 1963; then, applying the Miller standard of “probable” inclusion at defendant’s urging, the court set January-1, 1965 as the date when it became probable that the Woolstenhulme property would be taken. (See fn. 4, supra.)

Because defendant’s property lay immediately adjacent to the proposed lake, the trial judge might reasonably have found that this land was probably within the scope of the project from as early as the time in 1963 when the public first learned that some additional property would be needed for recreational facilities (cf. United States v. Crance (8th Cir. 1965) 341 F.2d 161, 165 ). The record makes clear, however, that during these early stages it was not known just how much of the property around the lake would be needed for public recreation, and, under these circumstances, the trial court could properly find that the probability of inclusion did not occur until the plans for the recreation sites became somewhat more definite around January 1, 1965. (Cf. United States v. 2,353.28 Acres of Land, etc., State of Fla. (5th Cir. 1969) 414 F.2d 965, 970-971 ; Calvo v. United States (9th Cir. 1962) 303 F.2d 902, 907-909 .)

Thereafter, in instructing the jury as to the proper determination of compensation, the trial judge directed the jury that it was not to “consider any enhancement that came about by virtue of public knowledge of this project for recreation purposes after [January] 1, 1965.” We conclude that this instruction did not permit the jury to award compensation for an increase in value to which the defendant was not entitled.

3. The trial court did not err in admitting evidence of sales which took place in the Lake McClure region in 1965 and 1966 as “comparable sales” under Evidence Code section 816.

The district contends that the trial judge erred in permitting defendant’s appraisal witness to support his opinion of the proper valuation of the land by presenting evidence of sales of nearby lands which occurred in 1965 and 1966. The trial court did find that these 1965 and 1966 sales reflected a “substantial enhancement” attributable to the recreational aspects of the Lake McClure project, but admitted them into evidence nonetheless, indicating that he would instruct the jury to eliminate improper enhancement. The district claims that sales which are found to reflect “substantial project enhancement” not properly shared by the condemned land, can never constitute “comparable sales” within the meaning of section 816 of the Evidence Code, and are thus inadmissible.

Section 816 of the Evidence Code provides in pertinent part that “[w]hen relevant to the determination of the value of property, a witness may take into account as a basis for his opinion the price and other terms and circumstances of any sale . . . [of] comparable property if the sale . . . was freely made in good faith within a reasonable time before or after the date of the valuation. In order to be considered comparable, the sale or contract must have been made sufficiently near in time to the date of the valuation, and the property sold must be located sufficiently near the property being valued, and must be sufficiently alike in respect to character, size, situation, usability, and improvements, to make it clear that the property sold and the property being valued are comparable in value and that the price realized for the property sold may be fairly considered as shedding light on the value of the property being valued.”

Given the inherent vagueness of this standard of “comparability,” appellate courts have recognized that “ ‘the trial judge . . . must be granted a wide discretion’ ” (County of Los Angeles v. Faus (1957) 48 Cal.2d 672, 678 [ 312 P.2d 680 ]) in determining the admissibility of sales sought to be relied upon as “comparable.” “[N]o general rule can be laid down regarding the degree of similarity that must exist to make such evidence admissible. It must necessarily vary with the circumstances of each particular case. Whether the properties are sufficiently similar to have some bearing on the value under consideration, and to be of any aid to the jury, must necessarily rest largely in the sound discretion of the trial court, which will not be interfered with unless abused.” (Wassenickv. Denver (1919) 67 Colo. 456, 464 [ 186 P. 533, 536 ]; see San Bernardino County Flood Control Dist. v. Sweet (1967) 255 Cal.App.2d 889, 905 [ 63 Cal.Rptr. 640 ]; People ex rel. State Park Com. v. Johnson (1962) 203 Cal.App.2d 712, 719 [ 22 Cal. Rptr. 149 ].)

Although the district does not deny that this broad discretion resides in the trial court, it does maintain that sales which are “substantially enhanced” can never properly be found to be “comparable sales,” because, assertedly by definition, such sales are not “sufficiently alike [the property to be valued] in respect to character, situation [or] usability. . . .” Section 816, however, does not establish criteria of “substantial” or “insubstantial” comparability, but rather requires the trial court to measure whether or not “the property sold” is “sufficiently alike” the property to be valued, by determining whether “the price realized for the property sold may be fairly considered as skedding ligkt on tke value of tke property being valued.” (Italics added.)

We recognize, of course, that in many, perhaps most, cases, a trial judge may find that sales of neighboring property which “substantially” reflect an enhancement value not properly shared by the condemned property, will not “shed light” on the value of the subject property, but rather will tend to confuse the issue if admitted into- evidence. In such cases the sales should, properly be excluded. We can conceive of a variety of situations, however, in which a trial court may reasonably find that such sales will “shed light” on the value of condemned land even though the sales reflect “substantial enhancement.”

In some cases, for example, a project will remain in the planning and construction stage for a great many years before a tract of land, originally designated for condemnation, is actually taken by the condemner. Although all sales in the neighborhood over that period may reflect “substantial project enhancement,” such sales may also reflect recent increases in land values attributable to other factors, such as other new public or private improvements or zoning changes, which the owner of the condemned land is entitled to have included in a consideration of the market value of his land at the time of taking. (See United States v. Miller (1943) 317 U.S. 369 , 373 and fn. 6 [ 87 L.Ed. 336, 342 , 63 S.Ct. 276 , 147 A.L.R. 55 ]; Urban Renewal Agency v. Spines (1968) 202 Kan. 262, 265-267 [ 447 P.2d 829, 831-833 ].)

Under these circumstances a trial court might reasonably conclude that the “substantially enhanced” sales could “fairly be considered as shedding light” on the value of the condemned property, since without the admission of such sales a landowner could not support his appraiser’s opinion of the increase in value attributable to these non-project factors. The conclusion is particularly viable if an expert appraisal witness can fairly estimate the amount of each of the enhanced sales prices which is attributable to “project enhancement.” In such a case, the trier of fact could subtract the amount of value which he finds to be due to project enhancement, and could then test the witness’s valuation of the condemned land against this “adjusted” sales price. Indeed, the trial court followed the latter procedure in the instant case: the defendant’s appraisal witness introduced evidence of other sales in the neighborhood and estimated the extent of “project enhanced value” at $50 an acre; the plaintiff contended, on the other hand, that in each of these sales, any amount over $125 an acre was attributable to project enhancement.

The district now argues, however, that in permitting defendant’s appraiser to isolate this “enhancement factor” in other, allegedly “comparable” sales, the trial court violated Evidence Code section 822, subdivision (d), which renders inadmissible “[a]n opinion as to the value of any property or property interest other than that being valued.” The district apparently reads section 822, subdivision (d), as precluding an appraiser, when referring to “comparable sales,” from explaining any adjustments that must be made in the “comparable sale” price in utilizing that sale as an indicant of the value of the property to be taken.

Such an interpretation of section 822, subdivision (d), however, goes considerably beyond the main purposes of that section and inevitably conflicts with the practical application of the entire “comparable sale” approach of section 816. Under the comprehensive statutory scheme relating to the evidentiary procedure for eminent domain proceedings enacted in 1961 (see, generally, Cal. Law Revision Com. Recommendations Relating to Evidence in Eminent Domain Proceedings (1960) [hereinafter cited as Law Rev. Com. Report]), appraisers, in relating their “opinion” as to the value of the property, are permitted to utilize a wide variety of valuation techniques, including “income capitalization” (Evid. Code, § 819), “reproduction” costs (Evid. Code, § 820) and comparative sale data (Evid. Code, §§ 816, 818). As the drafters of section 822, subdivision (d), indicated, in excluding “opinion” evidence as to the value of property other than the condemned property, the section simply attempts to avoid the host of collateral issues, and the consequent prolongation of eminent domain trials, that would arise if appraisers were permitted to testify, under these liberalized evidentiary rules, as to their “opinion” of the value of other property. (See Law Rev. Com. Report, p. A-8.) An appraiser’s testimony relating to adjustments to be made in “comparable sales,” however, does not normally raise collateral issues of great magnitude.

Moreover, the procedure of which the district complains is a most natural and, indeed, necessary component of the entire “comparable sales” approach sanctioned by section 816. It is a familiar statement that no two parcels of land are precisely equivalent; the property which is the subject of a “comparable sale” will always differ in some particulars from the property being valued. Commonly a “comparable sales price” will vary in some respect from an appraiser’s opinion of the condemned land’s “value”; when this happens, the appraiser will most naturally want to explain the distinguishing features between the property sold and the property to be valued, which he has taken into account in inferring the value of the land under consideration from the “comparable sale.” Moreover, even if the appraiser does not so testify on direct examination, he will frequently be questioned on cross-examination as to the relevant differences between the assertedly “comparable” parcel and the subject land. In response he will be compelled to disclose how he took these relevant differences into account in deriving his valuation figure. (See, e.g., City of Los Angeles v. Cole (1946) 28 Cal.2d 509, 518 [ 170 P.2d 928 ], overruled on other grounds in County of Los Angeles v. Faus (1957) 48 Cal.2d 672, 680 [ 312 P.2d 680 ].) Such inquiries are essential if the jury is intelligently to determine the weight that should be given to such “comparable sales” evidence. (See Law Rev. Com. Report, pp. A-50-A-51.)

Our courts have accepted this “adjustment” process as an integral ele ment of the “comparable sale” approach. In San Bernardino County Flood Control Dist. v. Sweet (1967) 255 Cal.App.2d 889 [ 63 Cal.Rptr. 640 ], for example, the court, in affirming the trial judge’s admission of “comparable sales” of property three to five miles distant from the subject property, stated: “The admissibility of testimony relating to comparable sales rests largely in the discretion of the trial court. [Citations.] In the present case, the court carefully considered the question of comparability and required the witness to adjust the sales prices to the date of value of the subject property. We find no abuse of discretion in the court’s ruling.” (255 Cal.App.2d at p. 905.) Likewise, in City of San Diego v. Boggeln (1958) 164 Cal.App.2d 1, 7-8 [ 330 P.2d 74 ], the procedure utilized by the court in the instant case was endorsed in the context of project “enhanced” comparable sales. (See County of Los Angeles v. Hoe (1955) 138 Cal.App.2d 74, 79-80 [ 291 P.2d 98 ]; cf. City of Gilroy v. Filice (1963) 221 Cal.App.2d 259, 271 [ 34 Cal.Rptr. 368 ]. See also United States V. Miller (1943) 317 U.S. 369, 380 [ 87 L.Ed. 336, 346 , 63 S.Ct. 276 , 147 A.L.R. 55 ]; State v. Wood (1969) 22 Utah 2d 317, 320-321 [ 452 P.2d 872, 874 ].)

The district also contends that even if “substantially enhanced” sales may be admitted under certain circumstances, such circumstances did not exist in the instant case; in other words, the district claims that the 1965 and 1966 sales were “noncomparable” as a matter of law and thus that the trial court’s admission of these sales constituted an abuse of discretion. Considerable testimony, however, attributed the rise in land values in the area to a substantial number of factors other than the Lake McClure project; the district’s appraisal witness, for example, conceded that the inflation of the mid-1960’s had affected the value of land around the state, and, as recounted earlier, the landowner’s witness cited a number of factors, including population growth and construction of freeways, as contributing to the increase in value. The trial judge could reasonably conclude that the 1965 and 1966 land sales might “shed light” on the effect of these factors on the property to be valued, particularly since, without the introduction of such sales, the jury would have been deprived of all “objective” market evidence on these matters. Under the circumstances, we conclude that the court did not abuse its discretion in permitting the witness to testify as to the challenged sales.

4. The trial court did not err in awarding defendant attorney’s fees in connection with a partial abandonment of the condemnation; it did err, however, in determining the scope of the abandonment.

Plaintiff raises one final issue on this appeal. The district contends that the trial court erred in awarding the landowner, Mrs. Woolstenhulme, $3,500 for attorney’s fees based upon a partial abandonment by the condemner. The award was made pursuant to section 1255a of the Code of Civil Procedure which provides that a condemnee shall be compensated for “reasonable costs and disbursements,” including attorney’s fees, which he incurs in preparing to defend a condemnation action which is later abandoned by the condemner.

In the initial complaint filed by the irrigation district in February 1966, the district sought to condemn (1) a fee interest in areas designated parcels 1, 2, 4 and 5 and (2) the cattle grazing and watering rights to 199.9 acres designated as parcel 3. Defendant and a predecessor had earlier sold parcel 3 to the district but had reserved the grazing and watering rights and, thus, the district’s intention in the initial complaint was to acquire the remainder of the complete fee interest in that tract. After this initial complaint was filed, defendant, through litigation, succeeded in rescinding her prior sale of parcel 3 to the district. The district, thereafter, in August 1967, filed an amended complaint, seeking condemnation of the fee interest of parcels 1 and 2 and 117 acres of parcel 3; this amended complaint dropped the demand for grazing and watering rights, and excluded parcels 4 and 5 completely. The trial court held that the amendment of the complaint constituted a partial abandonment, and awarded defendant an attorney’s fee of $3,500 based on money expended to defend parcels 4 and 5; and the grazing and watering rights of parcel 3.

The district does not, and could not properly, contend that the amended complaint did not constitute a “partial abandonment” entitling the landowner to attorney’s fees with respect to property and property rights omitted from the subsequent complaint. (County of Kern v. Galatas (1962) 200 Cal.App.2d 353, 356-357 [ 19 Cal.Rptr. 348 ].) The district, however, does raise two other objections to the $3,500 award.

First, the district, relying on the rule of Franklin-McKinley Sch. Dist. v. Lester (1963) 223 Cal.App.2d 347, 348-349 [ 35 Cal.Rptr. 727 ]; City of Los Angeles v. Welsh (1935) 10 Cal.App.2d 441, 443 [ 52 P.2d 296 ]; and City of Long Beach v. O’Donnell (1928) 91 Cal.App. 760, 761 [ 267 P. 585 ], contends that defendant was entitled to no award of attorney’s fees at all since, it is asserted, she had only a contingent fee contract with her attorney. Assuming, without deciding, that these cases correctly interpret section 1255a as precluding an award of attorney’s fees when those fees are purely contingent, we still cannot agree with the condemner that such fees should not have been awarded in the instant case.

Although the original contract between defendant and her lawyer provided only for a purely contingent fee arrangement, the attorney subsequently wrote his client stating that in the event of abandonment, the fee would be based on “reasonable charges” (see Cal. Condemnation Practice (Cont. Ed. Bar) pp. 18-19), and the trial court found that this second letter constituted a modification of the attorney-client fee agreement. The record contains substantial evidence to support a finding that defendant agreed to this modification of the fee contract, and therefore the trial court could properly find that the arrangement was no longer a purely contingent one. (Cf. Franklin-McKinley Sch. Dist. v. Lester (1963) 223 Cal.App.2d 347, 349 [ 35 Cal.Rptr. 121].) Thus, even under the authorities relied on by the district, the court could properly make an award under section 1255a.

Second, the district maintains that the trial court erred in characterizing the amended complaint as “abandoning” its instant demand for grazing and watering rights of parcel 3, and in awarding attorney’s fees related to the defense of those rights. We conclude that this contention has merit.

Section 1255a is designed to compensate a defendant for expenses incurred in anticipation of an eminent domain proceeding, when the condemner declines to carry the proceeding through to its conclusion. (Oak Grove School Dist. v. City Title Ins. Co. (1963) 217 Cal.App.2d 678, 698 [ 32 Cal.Rptr. 288 ].) By amending its complaint to seek a fee interest in 117 acres of parcel 3, while dropping its request for grazing and watering rights over the entire 199.9-acre tract, the district did abandon its efforts with respect to the 82.9 acres of parcel 3 omitted from the amended complaint. With respect to the 117-acre portion of parcel 3, however, the amendment did not constitute an abandonment of the initial claim for grazing and watering rights, but instead represented an enlargement of the original demand, seeking, in addition to the watering and grazing rights, all the other interests in the land which make up the fee simple estate. Thus, with respect to these 117 acres, the district did not fail to carry the proceeding through to conclusion; the services performed by the attorney with respect to that acreage were completely utilizable in the instant action. The court erred in viewing the district’s shift in position with respect to these 117 acres as an abandonment.

The abandonment was thus less extensive than understood by the trial court at the time it entered its cost award. The trial court is in the best position to determine how the reduced compass of the abandonment should affect the amount of the fee award and we believe that the proper disposi tion is to set aside the present cost order and remand this matter to the trial judge for recomputation.

We vacate the cost order and remand defendant’s motion for costs and disbursements to the trial court for recomputation in accordance with the conclusions expressed herein. In all other respects the judgment is affirmed. Plaintiff shall bear the costs of appeal.

Wright, C. J., McComb, J., Peters, J., Mosk, J., Burke, J., and Sullivan, J., concurred.

Several of the amici curiae in this matter have urged the court to address the issue of whether the depreciation of land values, resulting from the announcement of a public improvement is to be taken into consideration in computing just compensation. Although, of course, that issue and the enhancement issue presented by the facts of the three cases before us do show some correlations, we do not believe we should attempt to resolve the question of “project depreciation” (“project blight”) in the abstract. Most jurisdictions which have probed the problem do not follow identical rules with respect to project enhancement and project blight (4 Nichols on Eminent Domain (3d ed. 1962) § 12.3151[2], pp. 209-210), and several commentators have suggested that differential treatment may be the proper approach (see, e.g., Anderson, Consequences of Anticipated Eminent Domain Proceedings—Is Loss of Value a Factor? (1964) 5 Santa Clara Law. 35; Note, Recovery for Enhancement and Blight in California (1969) 20 Hastings L.J. 622 , 643-648). A major reason for a distinction between the two is that in the case of project blight, unlike enhancement, there is a danger that the government will announce the project in order to drive down neighborhood land values, and then attempt to take advantage of the depressed values when paying compensation for property it condemns. (See Uvodich v. Arizona Board of Regents (1969) 9 Ariz.App. 400 [ 453 P.2d 229, 234-235 ]; cf. United States v. Virginia Elec. & Power Co. (1961) 365 U.S. 624, 635-636 [ 5 L.Ed.2d 838, 848-849 , 81 S.Ct. 784 ].) In view of the additional complexities involved in the “blight” situation, we have concluded that before attempting to devise a general rule we should await a case presenting that matter directly. See generally, Note, Recovery for Enhancement and Blight in California (1969) 20 Hastings L.J. 622 . Some dispute has arisen over whether January 1, 1965 was the date at which the inclusion of defendant’s land became “definite” or just “reasonably probable.” At one point in the record the trial judge stated that “I am not going to apply a rule of certainty. I am going to use probability, apply the rule of probability.” Thereafter, when the judge set the date as January 1, 1965, he stated: “[T]his was a very fluid thing, but somewhere between the 29th of November, ’63 and December of 1965, this became pretty definite, that the Barrett Cove area and this property, or much of it, was going to be taken. And of necessity I must be a little bit arbitrary and I will make it January 1, 1965.” We believe the most reasonable interpretation of the record is that the January 1, 1965 date was reached by application of the “probability” standard. Actually 117 of the 189 acres involved in this action were known to be included in the project long before 1965, because those acres were to be actually flooded by the expansion of the lake; the recreation aspect concerned only 72 acres of the present parcel. Recognizing the difficulty the jury would have in understanding an extremely complex instruction submitted by defendant which drew this distinction, the district’s counsel agreed that the instruction could be modified to relate to the entire 189 acres. On this appeal both parties have treated the trial court’s finding as going to the inclusion of all of defendant’s property and, consequently, we adopt the same approach. The judge instructed the jury that: “You are not to take, to consider any increase in value after January 1, 1965—that is, related solely to the recreation. You may take enhancement into consideration—for example, what the experts have talked about, the natural increase in value of farm land, six or seven percent; any other factor of enhancement that may be in this case that you believe is applicable. . . . But you can’t consider any enhancement that came about by virtue of public knowledge of this project for recreational purposes after [January] 1, 1965. . . .” A11 of the early cases applying the Neale rule, did so to bar the inclusion of this type of “enhancement value.” (Sacramento So. R.R. Co. v. Heilbron (1909) 156 Cal. 408, 412 [ 104 P. 979 ]; City of Stockton v. Vote (1926) 76 Cal.App. 369, 404 [ 244 P. 609 ]; City of Pasadena v. Union Trust Co. (1934) 138 Cal.App. 21, 25-26 [ 31 P.2d 463 ].) In one passage in Neale the court did aver to this distinction between different kinds of “project enhancement.” After declaring that “benefit arising from the improvement” would be “inadmissible as a direct element of value,” the court observed: “It is possible that [the landowner] might get some benefit from [the project] indirectly. That is to say, the public knowledge of a proposed improvement might cause an actual demand in the market and a subsequent advance in the current rate of price. In such case it would be impracticable for a court to analyze the price and determine the proportion in which any particular element contributed thereto. The scales of justice do not balance quite so delicately as that. But aside from this indirect benefit ... it seems monstrous to say that the benefit arising from the proposed improvement is to be taken into consideration as an element of the value of the land.” (78 Cal. at pp. 74-75.) Although defendant reads this passage as firmly holding that “indirect enhancement" is a proper element of just compensation, we do not believe the decision can properly be interpreted as going that far. The quoted dictum does not declare that a landowner is entitled to this “indirect" benefit, but only that he might obtain this benefit because it would be “impracticable" for a court to analyze the price to eliminate this factor. In our view the discussion in Neale cannot be fairly said to have resolved the issue before us one way or the other. This analysis is also applicable to landowners who acquired the land prior to the public improvement. Although such owners have not paid out money in reliance on the project, they effectively have made an equivalent investment by retaining the land rather than selling it at the “enhanced price.” (See 1 Orgel on Valuation Under Eminent Domain (2d ed. 1953) § 98, p. 425.) Furthermore, if we were to ignore realities and were to require compensation up until the date of definite inclusion instead of the date of probable inclusion, we might effectively encourage the condemning authority to establish definite project boundaries quite hastily; we would thus discourage the government’s use of procedures, such as public hearings, which afford the public some direct participation in the planning and placement of such projects. Procedures permitting public participation inevitably delay the official pronouncement of the definite boundaries of a public project; these procedures might prove prohibitively costly if the government were required to pay for a rise in land values, not shared by the property likely to be condemned, that might occur during the course of public hearings. Courts have utilized a variety of linguistic tests in describing the requisite “certainty of inclusion” that is required before “project enhanced value” should be excluded. In the Miller case itself, the court, after initially declaring that the crucial question was whether the lands were “probably” within the project ( 317 U.S. at p. 377 [ 87 L.Ed. at p. 344 ]), later states that no “project enhanced value” should be considered if the lands were “within that area where they were likely to be taken for the project, but might not be . . .” ( 317 U.S. at p. 379 [ 87 L.Ed. at p. 345 ]) (italics added) (see also United States v. Crance (8th Cir. 1965) 341 F.2d 161, 163 (“might likely be acquired”); United States v. 172.80 Acres of Land, etc. (3d Cir. 1965) 350 F.2d 957, 959 (“probability of future inclusion”); Cole v. Boston Edison Company (1959) 338 Mass. 661, 666 [ 157 N.E.2d 209, 212 ] (“if it was contemplated . . . that . . . land in question would sooner or later be taken”) (original italics).) Despite this lack of uniformity or precision in terminology, however, most of the cases appear to exclude project enhancement whenever the court concludes that an informed owner could reasonably anticipate that the property might well be taken for the project. (See, e.g., United States v. Miller (1943) 317 U.S. 369, 377 [ 87 L.Ed. 336, 344 , 63 S.Ct. 276 , 147 A.L.R. 55 ] (enhancement excluded when “one probable [site]” for the project was marked out over defendant’s land); Shoemaker v. United States (1893) 147 U.S. 282 [ 37 L.Ed. 170 , 13 S.Ct. 361 ] (congressional act authorized acquisition of fixed acreage for park within larger area but did not fix boundaries of park; enhancement value excluded for all property within larger area).) In our view the “probability of inclusion” standard, utilized by the federal courts, expresses this concept adequately and in a readily comprehensible formula; the latter quality is certainly a most important one in this area, where the factual inquiries are invariably quite complex and frequently not susceptible to precise resolution. Accordingly, we believe that this standard is the appropriate one to be utilized in future cases. (See People ex rel. Dept. Pub. Wks. v. Arthofer (1966) 245 Cal.App.2d 454, 465 [ 54 Cal.Rptr. 878 ].) technically, it is possible that there may be some project enhancement of value even after this time, for some potential purchasers may conceivably be willing to pay more for such property in the hope, however remote, that ultimately the property will not be taken for the improvement. As we have explained earlier, however, any rise in value after this date is far more likely to be attributable to speculation upon the amount that the condemning authority will be compelled to pay. Because, as a practical matter, it would be impossible to determine the precise source of an increase in actual market value, and since those who purchase property after the date of probable inclusion voluntarily assume the risk of condemnation, we believe that the date of “probable inclusion” constitutes the most appropriate “cut-off” date for project enhancement. As stated in the text, the trial court conducted an inquiry into the date of “probable inclusion” and rendered a finding on that matter upon the agreement of both parties. We believe that, whether or not the parties so agree, such procedure should be followed in future cases. If the trial judge is precluded from making an early determination on this issue, he cannot properly determine which sales are sufficiently “comparable” to the condemned property to be admitted into evidence; furthermore, unless the trial judge is permitted to determine the appropriate “cutoff date,” we believe that, as a practical matter, it may be impossible to devise comprehensible instructions which explain to the jury which “enhanced value” is to be included in just compensation and which is to be excluded. We therefore conclude that the trial court, rather than the jury, should determine the issue of “probable inclusion.” The United States Supreme Court recently reached the same conclusion with respect to federal eminent domain proceedings. (United States v. Reynolds (1970) 397 U.S. 14, 20 [ 25 L.Ed.2d 12, 18 , 90 S.Ct. 803 ].) Initially, the trial judge inadvertently stated the date as October 1, 1965, but he immediately corrected the date to January 1, 1965, when counsel advised him of his slip. To the extent that “project enhanced” value is a proper element of the condemned land itself, other sales reflecting similar project enhancement may, of course, be considered comparable. Since we have concluded in the prior section that defendant was entitled to “project enhancement” until January 1, 1965, the condemner’s present objection is properly directed only at that element of the “comparable” sale prices reflecting project enhancement subsequent to January 1, 1965. Of course a trial court is not required to admit a proffered sale simply because an appraiser declares that he can isolate and eliminate all improper “enhancement” value. In every case it remains for the trial court, rather than the witness, to decide, from all the circumstances before it, whether a sale offered into evidence “may be fairly considered as shedding light on the value of the property being valued.” (See Los Angeles etc. School Dist. v. Swenson (1964) 226 Cal.App.2d 574, 583 [ 38 Cal. Rptr. 214 ].) In 1968, after the trial in this case, section 1255a was amended to codify the rule of the Kern case.