DUNIWAY, Circuit Judge:

This is an action to review a decision of the Secretary of Health, Education and Welfare denying an application for a period of disability and for disability insurance benefits under sections 216(i) and 223 of the Social Security Act as amended ( 42 U.S.C. § 416(i) and 423). The District Court entered summary judgment for the Secretary and the applicant (Flake) appeals. His arguments are, in essence, (1) that the decision of the Secretary is not supported by substantial evidence (§ 205(g)); and (2) that the Secretary applied an improper legal standard in finding the facts.

The case presents some important questions as to the proper construction of section 223 of the Act, as amended, and we therefore discuss it in some detail.

I. The facts.

Flake filed his application on October 10, 1963. It was denied by letter on January 13, 1964, and reconsideration was denied on June 27, 1964. He requested a hearing, which was held on November 16, 1964.

a. Flake's oral testimony.

In his application, Flake described his impairments as "coronary" and "osteoarthritis of the spine and extremities" and alleged that he became unable to work because of these impairments on June 10, 1963. At the hearing he was the principal witness. He was then fifty-eight years old, had a high school education, and had worked mainly as a retail clerk, although he had driven a truck for a short period. He had arthritis before entering the service in 1943 and had been on crutches for several months just before he was drafted. He was, however, able to walk into the induction center, and accordingly was drafted. The Navy in its infinite wisdom sent him to Farragut, Idaho, in the middle of winter, and shortly thereafter he was discharged for physical disability. He returned to Fresno, was unable to resume work as a truck driver, and after a short period of recovery, went to work as a clerk in a liquor store. He and his brother bought the store in 1946, operating it jointly until his heart attack in 1963. He worked only 3 days a week, but on those days he worked 12 hours. His arthritis continued to bother him, necessitating "boatloads of aspirin" and frequent shots.

On June 10, 1963, he suffered a heart attack and was hospitalized for forty-two days. After his release from the hospital, he spent several months in bed and then began to become more active. He tried to help his brother out at the liquor store on two occasions, but was unable to do so because he become nervous and upset, had chest pains, and his feet began to swell. His daily routine consisted of dressing himself (which took 45 minutes), taking pills, and light activity. He did drive an automobile on short trips.

He also complained of respiratory problems caused by recurring nasal polyps, and of generally short wind. He also had some hearing loss and dizziness. He complained about increasingly severe pain in his neck and spine, and knots in his hands from his arthritis.

Asked if he felt he could work, he replied: "I don't think I could. I would like to. I worked all my life, even when I was crippled up with arthritis and could hardly work * * * The last deal I had just threw me and I guess my equilibrium and my hearing, every three to four weeks it threw me out, and I have to stay in bed, sometimes weeks at a time. I get nauseated and sick to the stomach. I was picked up down town for being drunk and I don't drink. * * * I couldn't go back in the liquor store." He also felt that no one would hire him (the liquor store had been sold).

He had three shots a month, although he did not know what they were for. He continued to have chest pain, for which he frequently took nitroglycerin tablets. Asked if he could work if it weren't for the arthritis, he replied: I don't think so * * * I don't know. Maybe I have gotten to be cowardly. I have had so many friends that did the same thing, and they went back to work and two or three months later they were dead." His family doctor told him not to go back to work. He described work in a liquor store, saying that it involved a lot of tension because of drunks, appeals for credit, holdups, etc. It also involved being on one's feet and lifting cases of beer.

Finally, he described his equilibrium problems, for which he was treated by Dr. Snyder. He was taking pills that effectively controlled the problem.

Mrs. Flake then testified and corroborated her husband's story. When asked what she noticed to be wrong with her husband, she replied "Everything."

There was not other oral testimony.

b. The medical reports.

At the time of the hearing, a number of medical reports were before the Hearing Examiner. At the conclusion of that hearing, the examiner asked counsel for reports from various doctors who had treated Flake, but whose reports did not appear in the record. He also recommended that Flake go to an orthopedic specialist. He also wanted more in the way of objective medical findings from Flake's family doctor, who, he thought was "just being liberal on symptoms." The Hearing Examiner summed up his approach to a determination of disability:

"[W]e can't grant disability on the applicant's own statements of his condition. We have to have a medically determinable impairment. It means just that. It must be supported by medical evidence. I would say that the oral testimony is very strong, but the medical evidence is pretty weak."

As a result, Flake was examined by an orthopedist, and further medical reports were received and considered by the examiner. We summarize all of the reports that he considered.

First, there is a report from the Veteran's Administration of an examination conducted in 1948. The diagnosis was "chronic arthritis of the hands, feet and left shoulder and back."

Next, there are several reports and letters from Flake's family doctor, Dr. Hickman. The first is a letter to the Veteran's Administration dated February 25, 1963. This was about four months before Flake claimed to have become disabled. In it, the doctor stated that he had been seeing Flake since 1948, that Flake had a progressive history of osteoarthritis of the spine and extremities, complicated by myo-fibrositis, progressive diminished hearing, and that Flake was unable to work because of his arthritis. The second is a report made shortly after Flake's heart attack. The diagnosis was that Flake had suffered an extensive posterior-lateral myocardial infarction, from which he had recovered, and that he suffered from arthritis, equilibrium problems, and loss of hearing. The third is dated November 19, 1963. In it the doctor described Flake's condition in more detail. He concluded that Flake had coronary artery disease, coronary insufficiency, and anginal syndrome and that Flake's American Heart Association classification was "Class I" (no restrictions). He also diagnosed osteoarthritis of the spine and extremities with limitation of movement involving hands and knees. He described Flake's condition as static and said that "he is not able to carry on a gainful employment." The fourth report is dated on March 31, 1964. It described Flake's subjective symptoms as recurrent chest pains and fullness, inability to stand or work without distress, arthritis, diminished hearing, and recurrent nasal obstruction due to polyps. The medical findings were "possible" coronary heart disease, post-coronary syndrome and apprehension, arthritis in both hands and feet and entire cervical spine with limitation of movement and use and some deformity, nasal polyps, and diminished hearing. The EKG (electrocardiogram) showed good recovery from the heart attack and Flake was on anti-coagulant therapy. His condition was static. The doctor concluded that Flake should be considered disabled because of his "general physical condition." Dr. Hickman's last report is dated November 4, 1964, and states that he felt "this patient has physical incapacities that restricts [sic] him from carrying on any reasonable or strenuous, gainful employment."

Also in the record are two other letters written before Flake claims to have become disabled. One, dated February 28, 1963, is from a Dr. Bolkovatz, who had treated Flake since 1939. He stated that Flake had a progressive arthritic condition since that time and "[a]t present his arthritic condition has become so severe that he is unable to conduct his business." Dr. Bolkovatz also mentioned appellant's hearing problem. The other, dated March 1, 1963, was written by a Dr. Kass, who had treated Flake occasionally since 1950. He stated that Flake had had arthritis since that time, that it was getting progressively worse, and that he also had progressive loss of hearing. Dr. Kass says that "his disability prevents him from carrying on his normal activities in his work."

After the hearing, Flake was examined by an orthopedic surgeon, Dr. Downing. In his report, dated December 3, 1964, and after relating Flake's own version of his condition, Dr. Downing stated his findings: that Flake had a slight lower extremity limp, that motions were performed slowly, but throughout normal range, with some slight discomfort at the extremes, and that he noticed no swelling or deformities. He concluded that there were no definite positive findings. However, he wrote:

"Although this patient has no definite objective findings of arthritis at this time, his clinical history indicates that he has been suffering from a rheumatoid type of arthritis for many years and his general ageing plus association with this complaint is causing an incapacity for him to engage in gainful occupation. His complaints seem rational and can be accepted as true."

There is also a letter dated December 7, 1964, from Dr. Snyder, who saw Flake from 1960 to 1964. He diagnosed a Meniere's syndrome and possible coronary heart disease during the period before Flake's heart attack. When he saw Flake in 1964, he felt that Flake had suffered a myocardial infarct with resultant loss of ability to sustain great physical effort. He did not assess the possibility of rehabilitation, but felt that it was possible that Flake did indeed become short of breath with exertion. He did not comment on Flake's arthritis.

There is also a report from a Dr. O'Brien, a cardiovascular specialist, that adds nothing to Dr. Hickman's reports, except that he classed Flake as American Heart Association Class III-C (moderate restriction of ordinary activity) shortly after the heart attack.

Finally, there are two reports from Dr. Graveline, who examined Flake for the government. The essence of his first report, dated November 21, 1963, was that Flake seemed fully recovered from his heart attack (which the doctor felt was a coronary occlusion rather than a myocardial infarction) and had only mild arthritis. He also found borderline diabetes. These findings were based on his examination of Flake, the hospital records, and X-ray and laboratory findings. Dr. Graveline concluded that "[i]t does not appear that there is any disability at this time other than excess weight." Dr. Graveline again examined Flake on May 21, 1964, conducting a complete examination and having X-rays taken. Again, he was "unable to come up with any objective findings that would indicate that there is disability here." He said that Flake seemed fully recovered from his heart attack, and that the X-rays and the examination revealed only mild arthritis.

II. The decision.

The Hearing Examiner rendered his decision on January 25, 1965. After summarizing the evidence, he concluded that he was dealing with a conflict of opinion rather than objective medical fact. He stated that he was not bound by medical opinion as to disability, citing 20 C.F.R. § 404.1526, and gave such opinions weight only to the extent that they were supported by medical findings. He concluded:

"The Hearing Examiner does not doubt the honesty and sincerity of claimant in relating his symptoms and feelings of incapacity to engage in any kind of work. * * * However, Congress formulated a definition of disability which requires proof by medical evidence — impairment must be `medically determinable,' and this means not only as to the presence of an impairment, but as to its severity. For this reason claimant's own statements of his incapacity to work are not sufficient. * * * The burden of proof is on the claimant * * * and in the opinion of the Hearing Examiner, this burden has not been sustained. On the contrary, the best medical evidence negates a claim of incapacity to engage in any substantial gainful activity. No good reason appears, to the Hearing Examiner, why claimant cannot resume the type of work he was doing clerking in a liquor store, or at least he could certainly engage in some more sedentary occupation."

The Appeals Council approved the Hearing Examiner's decision and it thereby became the decision of the Secretary.

III. The Statute and the Regulations.

When Flake filed his application, and when the hearing was had and the decision was rendered, including that of the Appeals Board, section 223(c)(2) of the Act, as amended, defined "disability" as: "inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or to be of long-continued and indefinite duration. An individual shall not be considered to be under a disability unless he furnishes such proof of the existence thereof as may be required." (See 42 U.S.C. § 423(c)(2) (1964).)

On July 20, 1965, this language was amended by Pub.L. 89-97, § 303(a)(2) (79 Stat. 286, at 367) by substituting for the phrase "to be of long-continued and indefinite duration" the phrase "which has lasted or can be expected to last for a continuous period of not less than 12 months." This amendment is applicable to this case under section 303(f)(1) of Pub.L. 89-97 (79 Stat. at 368). Dean v. Gardner, 9 Cir., 1968, 393 F.2d 327, 328 n. 2. But it is by no means clear that such applicability can do Flake any good.

The Act was again amended by Pub.L. 90-248, Jan. 2, 1968, 81 Stat. 821. See 1 U.S.Code Cong. and Ad. News, 1967, pp. 923ff. The pertinent section is section 158. It shifts the definition of disability that we have quoted, without change, from section 223(c) to 223(d)(1)(A) and adds a new subdivision (3) to the new section 223(d) reading as follows:

"(3) For purposes of this subsection, a `physical or mental impairment' is an impairment that results from anatomical, physiological, or psychological abnormalities which are demonstrable by medically acceptable clinical and laboratory diagnostic techniques." (Id. at 983.)

It also shifts the second sentence of former section 223(c), quoted above in the body of this opinion, to section 223(d)(5), with only minor changes. Subsection (e) of section 158 of the amending Act makes these amendments also applicable to this case. Dean v. Gardner, supra.

Throughout the pendency of these proceedings, there were two other provisions of the Act that appear to be applicable. One is section 205(a) ( 42 U.S.C. § 405 (a)), which reads:

"(a) The Secretary shall * * * adopt reasonable and proper rules and regulations to regulate and provide for the nature and extent of the proofs and evidence and the method of taking and furnishing the same in order to establish the right to benefits hereunder."

The other is a part of section 205(g), dealing with judicial review. It provides:

"* * * where a claim has been denied by the Secretary or a decision is rendered under subsection (b) of this section which is adverse to an individual who was a party to the hearing before the Secretary, because of failure of the claimant or such individual to submit proof in conformity with any regulation prescribed under subsection (a) of this section, the court shall review only the question of conformity with such regulations and the validity of such regulations."

Pursuant to the authority thus conferred by section 205(a) the Secretary has adopted rather elaborate regulations relating to proof of disability. Those in effect during these proceedings appear at 20 C.F.R. § 404.1501-404.1539 (revised as of January 1, 1967). Particularly pertinent are sections 404.1502(a), 404.1510(a) as well as the remaining subdivisions of that section, 404.1511(a) and (c) relating to arthritis, and 404.1514 relating to impairment of the cardiovascular system. These sections specify in considerable detail the proof required. They are buttressed by section 404.1530. Several of these sections were amended in late 1967 (33 Fed.Reg. 15-18). In general the changes tighten the former requirements.

There is legislative history of new subsection 223(d)(3), quoted above, indicating that the Congress was concerned that the Secretary and the courts, particularly the latter, were being too liberal in finding disability, thereby jeopardizing the financial stability of the system. The Senate Report (Finance Committee) No. 744, Nov. 14, 1967, to accompany H.R. 12,080 (2 U.S.Code Cong. Admin. News, 1967, pp. 2834ff) discusses the problem (id. at 2880-2883). We find it significant that, in adopting section 223(d)(3) the Congress used the language that the Secretary had long had in his regulations, 20 C.F.R. § 404.1510, note 5, supra. It would appear that Congress felt that neither the Secretary nor the courts were following the regulations.

IV. Disposition of this case.

a. Sufficiency of the evidence.

We cannot say that the decision is not supported by substantial evidence (section 205(g)). It is true that all of the treating doctors found fairly severe arthritic impairment and substantial residual impairments resulting from Flake's heart attack. But the reports of Dr. Graveline and Dr. Downing, which appear to be based upon the kind of physical and laboratory examination contemplated by the regulations and now by the amended statute, indicate that there are few, if any, physical symptoms to support these findings. We note, too, that several doctors asserted that Flake was disabled at a time when he was still working in spite of his arthritic complaints, and that he continued to work until some months later, when he had his heart attack. Also, in the light of Dr. Graveline's findings, Dr. Hickman, who was Flake's regular physician, considerably modified his earlier diagnosis. To us, a fair reading of the record indicates that the Examiner did consider all of the evidence and chose to accept that of Dr. Graveline. He also observed Flake personally, stating:

"He has no outward appearance of disability. He looked healthy and strong, was mentally alert and intelligent."

b. The correct legal standard.

Even though the findings be supported by substantial evidence, the decision should be set aside if the proper legal standards were not applied in weighing the evidence and making the decision.

In one respect, it is clear that the Examiner and the Appeals Council did not apply the presently applicable legal standard. It did not then exist. We refer to the 1965 amendment which we have quoted earlier in this opinion. What the result might have been under the "not less than 12 months" standard we cannot say. It is undisputed that Flake did suffer a heart attack serious enough to put him in the hospital for 42 days, that Dr. Hickman (the treating general practitioner) and Dr. O'Brien (the consulting cardiologist) as well as Dr. Snyder, diagnosed a myocardial infarct (Dr. O'Brien says "severe"), that Flake was bedridden at home for some three months, and that he recovered slowly thereafter. Surely, this is a "disability" within the meaning of the statute and regulations, if likely to continue for 12 months. See especially 20 C.F.R. § 404.1502, note 4, supra. Confronted with the "not less than 12 months" standard, the Hearing Examiner might well have found a period of disability.

It is also true that neither the Examiner nor the Appeals Council had before them the January 2, 1968, amendment, section 223(d). We have already discussed its effect. We would only add that we doubt that it was designed to restrict the medical evidence to "objective" or "physical" symptoms. We used the word "objective" in discussing section 223(d) in Ryan v. Secretary, 9 Cir., 1968, 393 F.2d 340. But the statute does not use it. Webster's New International Dictionary, 2d Ed., gives this definition:

"6. Med. Perceptible to persons other than the patient; — of symptoms."

The statute refers to "medically acceptable clinical * * * diagnostic techniques." Stedman's Medical Dictionary, 20th ed., defines "clinical" as:

"2. Denoting the symptoms and course of a disease as distinguished from the laboratory findings of anatomical changes."

We venture to suggest that there may be people who are really disabled, and can be found so by medically acceptable clinical diagnostic techniques, even though laboratory techniques do not support the diagnosis. See Mark v. Celebrezze, 9 Cir., 1965, 348 F.2d 289, and cases there cited; Page v. Celebrezze, 5 Cir., 1963, 311 F.2d 757. Not all subjective complaints of a patient are accepted by a doctor. But one skilled in the art may well be able, by medically acceptable clinical techniques, to sort them out, to decide which to believe, and to make a diagnosis accordingly. In this case, several doctors, in spite of a lack of "objective" symptoms, believed Flake's complaints and came to a diagnosis on that basis. See particularly the report of Dr. Downing, quoted supra. It is arguable that the Hearing Examiner felt bound to reject those diagnoses. We are not at all sure that the amended statute or the regulations require this. It may be, however, that Page v. Celebrezze, supra, and cases like it, are those to which the Senate Committee was referring. See note 9, supra.

The Hearing Examiner may also have applied an incorrect standard to the correlative burdens of the claimant and the Secretary in establishing disability. He said, "at least * * * [the claimant] could certainly engage in some more sedentary occupation [than clerking in a liquor store]", implying that the burden was upon Flake to establish both that he was unable to return to the liquor store and that he was unable to engage in any more sedentary occupation. Such a standard would conflict with our decision in Rosin v. Secretary, 1967, 379 F.2d 189. Whether that decision survives the 1968 amendments (see § 223(d) (5) of the Act; S. Rep. No. 744, supra, at 2881-2883; Dean v. Gardner, supra) and what the proper burden of proof should be under those amendments are questions we think should be decided by the Secretary on remand.

Nothing that we have said is to be taken as a definitive construction of the 1965 or 1968 amendments to the Act, or of the regulations. We only suggest some problems that appear to us to lurk in them. The primary responsibility for interpreting and applying both is on the Secretary. Accordingly, we vacate the judgment, and direct that the matter be remanded to the Secretary for further hearing and decision. Both parties should be permitted to present additional evidence if they so desire.

Vacated and remanded.