These appeals by the United States from District Court judgments approving valuation awards of condemnation commissioners present the question whwher the reports of the commissioners were sufficiently detailed as to findings of fact, and in giving the basis on which they were bottomed to permit adequate review by the district court, and thereafter by this Court.
These condemnation suits are a part of a program of land acquisition for the Walter F. George Lock and Dam project on the Chattahoochee River in Georgia and Alabama. The three series of tracts involved are all tracts of ordinary farm, timber and pasture land. While the government requested a jury trial, it does not now contend that the submission of the issues to commissioners by the trial court constituted reversible error. Neither does the government, on the record before us, assign any specific error in the hearings before the commissioners as a basis for attacking the judgment of the district court in affirming the report of the commissioners. Instead, the government takes the position that the reports of the commissioners are in such general and vague terms as to make it impossible for the district court or for this Court on appeal to determine whether the commissioners’ ultimate conclusions of value were based on legally cor*777rect principles or on legally sufficient evidence.
The appellees here, first, claim that the Court of Appeals cannot take notice of the government’s appeal, because it does not expressly contend that the awards are excessive. They take the position, therefore, that any claim by the government that there was error in the form or substance of the commissioners’ reports falls within the “harmless error” rule of the Federal Rules of Civil Procedure, Rule 61 F.R.C.P. We think this contention is without merit. It is apparent from the findings of the commissioners that the awards made by them and approved by the trial court are substantially in excess of the amounts deposited in court upon the declaration of taking. It thus appears that if errors occurred in the proceedings below, as a result of which judgments were awarded in excess of the amounts contended for by the United States, such errors would not be harmless in the sense of the Rule.
In general, the report of the commissioners recited the substance of the valuation testimony given by the several witnesses tendered by the landowners, on the one hand, and by the government on the other, and they made ultimate findings of the market value of the property taken and of severance damages to lands not taken and make certain specific findings as to the value of easements and fences taken, or the cost of fencing the remaining tracts. The reports did not in any manner whatever indicate which evidence the commission credited and which evidence it discredited. The reports gave no indication as to the degree to which it based its findings upon those opinions that were based on knowledge of comparable sales, nor did they give any indication as to whether indicated sales were truly comparable. The reports did not indicate to what extent it gave credence to the opinions of witnesses who, according to the summary of evidence given in the reports themselves, had little or no familiarity with the ordinary ingredients that are generally considered by the courts to be required to support an opinion of value in a coa-demnation case.
Part of the basis for this7 Court’s repeatedly stating that the trial of condemnation valuation issues is typically for a jury, and the appointment of commissioners is proper only in the exceptional case, United States v. Buhler, 5 Cir., 254 F.2d 876, United States v. Leavell & Ponder, Inc., 5 Cir., 286 F.2d 398, 408, United States v. Buhler, 5 Cir., 305 F.2d 319, 331, is that where a trial is had before a jury, the trial judge is charged with the responsibility of determining the qualifications of so-called expert witnesses and of others who undertake to express opinions as to land values and to determine initially whether so-called “comparable sales,” about which witnesses propose to testify, are sufficiently comparable to justify their consideration by the fact-finder, and because, further, the trial judge is able, by correctly charging the jury, to point out the defects and weaknesses in the testimony of interested parties to the proceedings, such as the owners of the land involved, and to stress the importance, as courts have always done, Baetjer v. United States, 1 Cir., 143 F.2d 391, 397; International Paper Co. v. United States, 5 Cir., 227 F.2d 201, 208; United States v. Leavell & Ponder, Inc., 5 Cir., 286 F.2d 398, 407, cert. denied 366 U.S. 944, 81 S.Ct. 1674, 6 L.Ed.2d 855, of opinion evidence based on comparable transactions. In a trial to a jury under such supervision by a trial judge, it can well be understood why a general verdict of value, plus a general verdict of severance damages can suffice, whereas a hearing before a commission must result in findings much more detailed than a general verdict.
The Courts of Appeals of the several circuits are not of a uniform mind as to this, but we find ourselves fully in accord with the reasoning of the Court of Appeals for the Fourth Circuit in United States v. Cunningham, 4 Cir., 246 F.2d 330, 333, where it is said:
“The very reasons which justify the appointment of the commission,. *778however, demonstrates the inadequacy of the commission’s report. The justification of the appointment is the variety and complexity of the matters to be considered on the question of valuation and the ' importance of having these adequately set forth in a report so that they may be subjected to the scrutiny of the District Court and of this court upon i-eview and the proper principles of valuation applied to them. Any adequate review of the facts or of the legal principles followed in basing valuations on the facts is defeated if a report by the commission is of such a character that it amounts to no more than a general verdict by a jury. The verdict of a jury of twelve men may reasonably be dispensed with if commissioners make a report which furnishes an adequate basis of review by the trial judge and the appellate court, but not if the report furnishes no such basis. Just as a judge in a trial without a jury is required to make adequate findings so that his conclusions may be reviewed by the appellate court so a master, in an action to be tried without a jury, is required to make findings of fact so that his conclusions may be adequately reviewed by the trial judge, who is required to accept them ‘unless clearly erroneous’ (Rule 53(e) (2)), and this practice with respect to the report of a master is prescribed by Rule 71A (h) with respect to reports of commissioners in condemnation proceedings.”
This Court has expressly approved this language of the Fourth Circuit in United States v. 2,477.79 Acres of Land in Bell County, 5 Cir., 259 F.2d 23, where on page 29, we said:
“We find ourselves in agreement with the Government’s position that the findings of the commissioners are wholly inadequate and that the judgment must be vacated and the cause remanded for proper findings and a judgment based thereon. From the report, so styled, of the commissioners nothing appears except a recital of their appointment, a statement that a hearing was had, and the commissioners’ conclusions as to values. As examples of the deficiencies in the findings it may be noted that nothing is found as to how the commissioners resolved the conflicts in the testimony, no findings appear as to the uses of the land, particularly Tract 805, and no determination is made as to benefits. Without explicit findings the trial court cannot adopt or reject the findings or adopt some and reject others. Without adequate findings this Court does not have before it a record which permits of a review of the district court’s adjudication. United States v. Buhler, supra; United States v. Cunningham, 4 Cir., 1957, 246 F.2d 330. [Emphasis added.]”
We recognize that the view we take of this matter is at variance with that of the Court of Appeals for the Tenth Circuit as expressed in the decision of that Court in United States v. Merz, 10 Cir., 306 F.2d 39. Our view in this respect is in accord with that of the Court of Appeals for the Ninth Circuit, which has recently reversed judgment of a trial court in California which expressly stated that in a condemnation case, the Commission’s finding “may be as general as the verdict of a jury, and have the same effect.” In its judgment reversing this decision the Court of Appeals in United States v. Lewis, 9 Cir., 308 F.2d 453, said:
“Upon this basic difference we agree with the principles expressed in Cunningham. The district judge is not sitting as the presiding judge in a jury-tried case, but as a reviewing court. He has not heard the evidence nor supervised its admission and thus is in no position to view the commission’s report simply as a jury verdict. If the court is intelligently to perform a function of review, it must be able to ascertain *779whether, in arriving at its value judgment, error was committed by the commission, either in the resolution of factual disputes or in the application of principles of valuation. There must be a sufficient disclosure to the reviewing court to enable it to understand what it is that has been decided. As stated by Mr. Justice Cardozo in United States v. Chicago, M., St. P. & P. Railroad, 1935, 294 U.S. 499, 510-511, 55 S. Ct. 462, 79 L.Ed. 1023:
“ ‘We must know what a decision means before the duty becomes ours to say whether it is right or wrong.’ ”
Although admittedly the scope of review, either by the trial court of the commissioners’ findings of valuation or by this Court on the trial court’s judgment, is restricted in the sense that a reversal is to be had only if the order under review is clearly erroneous, this function cannot be adequately performed by the reviewing court unless the fact finder makes it plain what the basis of its decision was. We do not say that every contested issue raised on the record before the commission must be resolved by a separate finding of fact. We do say, however, that there must be sufficient findings of subsidiary facts so that it will appear to the reviewing court that the ultimate finding of value was soundly and legally based. It is too obvious to require argument that in determining market value the best test is what the same or similar property is selling for in the locality at or near the day of taking. Thus, it is universally recognized that the best test of market value is the data concerning comparable sales. On the record before us, the commission speaks of comparable sales, but there is no finding or expression of opinion as to whether the sales sustain a value of $100 per acre, for instance, in the case of one of the tracts, as found by the commission, or whether this value represents merely a scaling down by the commission of an expression of an opinion by. others whose opinion of value may have been based on some such theory as that expressed by one of the witnesses, who-said:
“I arrived at it like I’d price-mine, just like it is; that’s what it would take to buy mine adjoining-Hoke’s.”
If this is all that the record shows as to-this neighbor’s qualifications to express, an opinion of value of the land, then such opinion would obviously have no probative value.
This Court has held that in an appeal from a judgment of the trial court where the valuation finding has been made by commissioners, it is the function of the trial court to determine whether the commission’s findings are to be approved or are to be reversed under the clearly erroneous standard. United States v. Twin City Power Company of Georgia, 5 Cir., 253 F.2d 197, 204. It is then the function of this Court to determine whether the district court’s disposition of the commission’s findings was clearly erroneous.
In order that the district court may perform its function, it must be able to determine whether the commission adopted the correct legal principles, and whether the evidence before the commission met the standard of substantiality to withstand a reversal by the district court. As we said in United States v. Leavell & Ponder, Inc., 5 Cir., 286 F.2d 398, at page 406, “The figure arrived at by the commissioners would much better have been supported by subsidiary findings of fact. * * * Without adequate subsidiary findings it is impossible for this Court to test the correctness of the elements of which it is the product or the sum.”
Furthermore, while we do not even suggest the need for long findings or long reports merely for the sake of length, there is much to be said for the view that commissioners like trial judges may be expected to give more careful consideration to the subsidiary facts and the legal *780principles involved if they are required to be stated in the report. The language of the opinion of the Court of Appeals for the Second Circuit, in United States v. Forness, 2 Cir., 125 F.2d 928, at page 942 is here apposite:
“It is sometimes said that the requirement that the trial judge file findings of fact is for the convenience of the upper courts. While it does serve that end, it has a far more important purpose — that of evoking care on the part of the trial judge in ascertaining the facts. For, as every judge knows, to set down in precise words the facts as he finds them is the best way to avoid carelessness in the discharge of that duty: Often a strong impression that, on the basis of the evidence, the facts are thus-and-so gives way when it comes to expressing that impression on paper.”
In order that the commissioners’ reports may meet the standards here prescribed, the judgments are reversed and the cases are remanded to the trial court for resubmission.