Appellee, hereafter sometimes referred to as Strolee, was the victor in a patent infringement suit brought in the District Court for the Southern District of California against appellant, hereafter sometimes referred to as Welsh. The patents held and sued upon by Strolee are United States Letters Patent Nos. 2,728,580 and 2,798,730, otherwise known as the Preis-ler patent or stroller and the Smith patent or stroller respectively. The Preisler patent teaches a collapsible baby stroller and the Smith patent one that is not only collapsible but also adjustable for lying down, sitting up and intermediate positions. Upon this appeal, Welsh concedes that if the patents are valid, it has infringed them. It urges, however, that *510 the patents are not valid and, secondarily, that the trial court committed prejudicial error in regard to the exclusion of certain evidence.

The Preisler Patent

As we read appellant’s brief, it raises two contentions in regard to the invalidity of this patent: 1.) That the patent was fully anticipated by the prior art, see 35 U.S.C. § 102 (a), consisting of baby strollers manufactured by the Kuniholm Company and designated in the record as Exhibits E, H, L, R, P, and AP, and 2.) that if this prior art did not fully anticipate the Preisler patent, the latter is invalid for lack of invention. See 35 U.S.C.. § 103; Leishman v. General Motors Corp., 9 Cir., 1951, 191 F.2d 522, 530 ; Rohr Aircraft Corp. v. Rubber Teck, Inc., 9 Cir., 1959, 266 F.2d 613, 618-619 . 1

Both of these contentions necessarily raise a question of fact — what was the prior art at the time appellee’s devise was conceived and reduced to practice? See 35 U.S.C. § 102 (g). The trial court’s resolution of this question appears in the record as Findings of Fact 24 and 25, which we reproduce here for easy reference.

We think these findings are too broad and conclusory. They do not reveal the “basic facts on which the District Court relied.” Dalehite v. United States, 1953, 346 U.S. 15 , 24 note 8, 73 S.Ct. 956, 962 , 97 L.Ed. 1427 . Apparently the court found that not only the device taught by the Preisler patent but also the Kuniholm strollers, exhibits H, L, R, P, Y and AA, had been reduced to practice before the filing date of the Preisler application, February 9, 1953. Otherwise, Finding of Fact 24 would serve no pui’pose; the Kuniholm strollers there mentioned would have been included in Finding 25. And although the court clearly found that appellant had failed to prove that the Kuniholm strollers were reduced to practice prior to the date of reduction to practice of the Preisler device, the essential facts leading to this conclusion are unexpressed. These missing findings are of the utmost significance. Without them we cannot determine whether the trial court held that appellant had failed to prove prior use because its evidence was insufficient to show the dates of reduction to practice which it claimed for the Kuniholm strollers or because appellee’s evidence of the date of reduction to practice claimed for the Preisler stroller — approximately fifteen months before the filing of Preisler’s patent application — was sufficient to establish the latter’s priority regardless of appellant’s evidence. We have not been told upon what dates each of the Kuniholm strollers and the Preisler device were, in the trial court’s view, reduced to practice. And even if we were made aware of such conclusions, we could not review the court’s interpretation of the meaning of “reduction to practice,” assumedly a legal standard, because the court has made no factual findings as to the events which transpired in regard to the various strollers during the period prior to February 9, 1953. We could not say whether or not these events constituted a reduction to practice because we do not know what the events were.

*511 The conclusion reached by the District Court in regard to priority appears as Finding 24. We can only imagine the court’s view of the basic facts. Conceivably, the District Court thought that the Preisler stroller had been reduced to practice in October, 1951, since the evidence indicates that at no other time prior to February, 1953 did anything happen which could possibly be considered as a reduction to practice of the Preisler device. Yet testimony does suggest that something happened in October, 1951. We could further assume that the dates of deduction claimed by appellant for the various Kuniholm strollers were accepted by the trial court except for the date claimed for Exhibit H. All other dates were later than October, 1951. As to Exhibit H, we could assume that the trial court concluded that the evidence offered as to the date of reduction to practice was insufficient to accomplish the end to which it was addressed. We could assume all these findings and then proceed to test them by the “clearly erroneous” standard prescribed by Rule 52(a) of the Federal Rules of Civil Procedure, 28 U.S.C. We could also assume that the trial court took another course, that it thought none of the Kuniholm strollers were reduced to practice on the dates claimed by appellant. We could then subject this assumed finding to the “clearly erroneous” test. If the evidence supported all of these assumed findings, regardless of their inconsistencies, then no matter how the District Court arrived at its conclusion, we could not overturn it. But we think it is the duty of the District Court to find the facts and not to leave to us the heavy chore of reviewing sundry, contradictory assumptions any of which could have led to the eonclusory statements misnamed Findings of Fact in the present record.

“Findings of fact are required under Rule 52(a), Federal Rules of Civil Procedure * * *. The findings should be so explicit as to give the appellate court a clear understanding of the basis of the trial court’s decision and to enable it to determine the ground on which the trial court reached its decision. * * * This court is not the trier of facts, nor does it substitute its own judgment for that of the trial court.” Irish v. United States, 9 Cir., 1955, 225 F.2d 3, 8 . See also Dalehite v. United States, supra; Hycon Mfg. Co. v. H. Koch & Sons, 9 Cir., 219 F.2d 353 , certiorari denied 1955, 349 U.S. 953 , 75 S.Ct. 881 , 99 L.Ed. 1278 ; Cross v. Pasley, 8 Cir., 1959, 267 F.2d 824, 826 ; Graham v. United States, 9 Cir., 1957, 243 F.2d 919, 923 .

In the instant case, we are wholly unable to determine how the trial court reached its conclusions regarding priority of invention between the various Kuniholm strollers on the one hand and the Preisler device on the other.

The inadequacy of the District Court’s fact findings also permeates its conclusions concerning anticipation and invention. Findings of Fact 27, 28 and 29 read as follows:

We do not think it plausible to read Finding 27 as assuming, contrary to the express conclusion reached in Finding 24, that the various Kuniholm strollers comprised part of the prior art, and as rul *512 ing, given such an assumption, that the Preisler device was significantly different from the Kuniholm products. Finding 27 says no more than that the Kuni-holm strollers did not anticipate the Preisler device. To have anticipated in this case, the Kuniholm strollers would have had to have been part of the prior art. For Finding 27 is apparently predicated upon Finding 24; the Kuniholm strollers are not anticipatory because they were not prior. As we have previously noted, the factual basis upon which the trial court rendered its conclusions concerning priority or the identity of the prior art is unrevealed to us. In consequence, we cannot discern the factual basis for the court’s conclusions as to anticipation. Again we are without findings concerning the basic facts which led to the trial court’s decision.

Findings 28 and 29 suffer from a similar deficiency. They clearly represent conclusions of law; that is, an application of the statutory standard of invention. See generally, Chin, The Statutory Standard of Invention; Section 103 of the 1952 Patent Act, 3 Patent, Trademark and Copyright J. of Res. and Educ. 317 (1959). As Judge Pope noted in his penetrating concurrence in Bergman v. Aluminum Lock Shingle Corp., 9 Cir., 1957, 251 F.2d 801 , 809, the presence or lack of invention will frequently turn upon findings of fact. These findings relate to the differences between the claimed invention and the prior art. In the present case such differences appear in Findings of Fact 7, 8, 9 and 10 which in effect describe a problem which existed in the prior art and relate the manner in which the claimed invention solved that problem. These findings, however, indicate the differences between the Preisler device and the “prior art” while Findings 28 and 29 apply the standard of invention to a comparison between the Preisler stroller and the listed Kuniholm strollers. The differences between the Preisler device and the prior art are not necessarily the differences between the Preisler device and the Kuniholm strollers, for under Finding 24 the District Court held that the Kuniholm strollers were not part of the prior art. We are thus left in a quandary concerning the trial court’s view as to what the differences are between the Kuniholm strollers and the Preisler strollers. We do have the strollers before us as exhibits, but there was expert testimony given at the trial, some of it quite pertinent to the question of invention. Belief or disbelief of such testimony is, in this case, of more than passing import. We are therefore not in the same position as the trial court, which could judge the credibility of the witnesses. And*because the trial court has failed to tell us how in its opinion the Preisler device differs from the assortment of Kuniholm strollers, we are without adequate findings of fact to review the question of invention.

Thus, we cannot assume the priority of the Kuniholm strollers and address ourselves to the questions of anticipation and invention, remanding for further findings as to priority only in the event that we were to conclude that, given such priority, there was anticipation by or a lack of invention over the Kuniholm devices. We are as handicapped in regard to anticipation and invention by the absence of adequate findings of fact as we are in regard to the question of priority. Or to put it another way, we cannot review the questions of invention or anticipation in this ease without considering the identity of the prior art.

Pursuant to the cases cited on page 511, supra, the judgment of the District Court insofar as it pertains to the validity of the Preisler patent is vacated, and the case is remanded for further findings of fact as called for in the preceding paragraphs.

The Smith Patent

The evidence shows and the District Court found, see Finding of Fact 30, that the Smith device is no more than the Preisler stroller enhanced by an adjustable back rest and an adjustable foot rest. When the occupant of a Smith stroller is in a sitting position, the back rest can be adjusted to a vertical posi *513 tion to support his back. When the child is lying down, the back rest can be adjusted to a horizontal position, again supplying support for his back. The foot rest is adjustable in that it rests beneath the seat to provide support for the occupant’s feet while he is seated, yet may be raised to the same plane as the seat and the back rest when the occupant is lying down. In this second position the foot rest will support the child’s legs. Welsh claims that the device taught by the Smith patent is invalid for lack of invention over the prior art comprised of the Dann, Maurer and Young patents and the Kuniholm strollers designated in the present record as Exhibits P and AB. The District Court found as facts:

It would appear from these findings that the District Court determined that the Dann, Maurer and Young patents and the Kuniholm stroller, Exhibit P, were all part of the prior art over which the Smith device represented an invention. As far as Exhibit AB is concerned, the District Court concluded that it was not part of the prior art but once again failed to reveal the reasons impelling such a conclusion. And, significantly, appellee claims that the Smith stroller' was reduced to practice more than a year prior to the patentee’s patent application. As with the various Kuniholm *514 strollers alleged to have been invented prior to the Preisler patent, we are confronted with insufficient findings of fact from which to determine the identity of the prior art, in this instance, the priority between the Smith device and Exhibit AB.

As we have mentioned, the findings of the District Court imply that Exhibit P was part of the prior art over which, in the District Court’s view, the Smith stroller constitutes an invention. Yet there is no express finding which flatly states that Exhibit P was as a matter of fact prior to the Smith device. Indeed, appellee claims on appeal that Welsh failed to prove the date of reduction to practice of Exhibit P as well as those of other Kuniholm strollers. The trial court does state clearly those differences between Exhibit P and the Smith stroller which led it, the court, to conclude that the latter was an invention over the former, and, assuming the priority of Exhibit P, we are afforded adequate fact findings upon which to review the trial court’s application of the standard of invention, something which we were unable to do on the basis of the fact findings regarding the Preisler patent.

If we thought that there was no question but that the Smith device did achieve the heights of invention when compared to Exhibit P, we would be inclined to deal with the problem on the assumption that the District Court did find that Exhibit P was part of the prior art, remanding only insofar as is necessary to bring out the basic facts concerning Smith’s priority over Exhibit AB. We think, however, that the question of invention over Exhibit P, when that stroller is considered together with the prior patents relied on by Welsh, constitutes a problem of some difficulty, and we prefer not to address it upon assumed findings concerning the priority of Exhibit P. A proper resolution of this problem should, in our opinion, await more specific findings of fact by the trial court. We would like to know the court’s opinion as to the dates of reduction to practice of the Smith stroller and of the Kuniholm devices, Exhibits P and AB, and, further, what the facts are which support the conclusions as to the various reduction to practice dates. Only when such findings are before us will we have an adequate record upon which to review the trial court’s conclusions as to the identity of the prior art, and, in consequence, the court’s ruling that the Smith patent is not void for lack of invention.

The judgment of the District Court is vacated and the case is remanded for further findings of fact as called for in this opinion and for such other treatment as the trial court may deem proper.

. Appellee’s argument that tills court should not consider the question of anticipation because of a purported deficiency in appellant’s Statement of Points on Appeal is -without merit. In the context of the instant case, appellant’s Statement constitutes a sufficient compliance with our rules.