Harley T. TALBOT, Plaintiff-Appellant, v. Margaret M. HECKLER, Secretary of Health and Human Services, Defendant-Appellee

STEPHEN H. ANDERSON, Circuit Judge. Harley Talbot applied twice for Social Security disability insurance benefits, first on June 9, 1982, and again a year later on June 13, 1983. Both applications were denied by the Social Security Administration (“SSA”), and Talbot (“claimant”) appealed the second denial to an administrative law judge (“AU”). His appeal was denied by the AU, and the Appeals Council of the Secretary of Health and Human Services affirmed the denial. The claimant then sought review in federal district court, where the administrative actions were upheld. The claimant now appeals from the district court’s adverse ruling. We reverse, on the basis that substantial evidence does not support the AU’s decision that the claimant could engage in a full range of light work. BACKGROUND The claimant is a fifty-six-year-old man who has not been gainfully employed since October 10, 1981. He has a recent history of heart disease, lung disease, and liver disease. He attended public schools through the eighth grade and received the *1458 equivalent of a twelfth grade education via correspondence courses. For over thirty years, he worked on a variety of jobs in highway, bridge, dam, and pipeline surveying and construction, work requiring medium to heavy exertion. Among other jobs, he held the position of “public works inspector,” “construction superviser,” and “field engineer.” Over the past several years, Mr. Talbot has experienced an increasing number of health problems, including blackouts and dizzy spells. He complains of chest pain, back pain, and shortness of breath upon exertion. His treating physicians have found objective evidence of atrial fibrillation (irregular contractions of the atrium of the heart) and atherosclerotic heart disease (heart disease involving degeneration of the arteries). During the claimant’s initial application process for Social Security insurance benefits in mid 1982, treating physician Dr. Beryl R. McCann reported that the claimant was suffering from atrial fibrillation and congestive heart failure subsequent to pneumonia, for which he was hospitalized in late 1980. Dr. Alice D. Cox, who examined the claimant in mid-1982 at the request of the SSA, did not find congestive heart failure but did confirm atrial fibrillation probably secondary to atherosclerotic heart disease. After ordering tests of pulmonary function, she also concluded that the claimant suffered from chronic obstructive pulmonary disease. A few months later, during the reconsideration of the claimant’s first application, Dr. McCann reported that the claimant was basically physically incapacitated, citing his heart problems, hypertension, intermittent chest pain, and blackout episodes. By the time of the claimant’s second application for benefits, a report was available from treating physician, Dr. Billy Joe Matter, who had examined the claimant in January of 1983 and diagnosed both lung and heart disease. In addition, Dr. Matter reported signs of liver disease, concluding that the claimant could not perform ordinary manual labor and could work only in a “limited sedentary situation.” Another treating physician, Dr. Wallace Byrd, reported in June 1983 that the claimant was permanently disabled as a result of his heart disease, hypertension, liver disease, unstable angina, and intercurrent infection. Dr. Byrd continued to see the claimant over a period of ten months from June 1983 to April 1984. In a hospital discharge summary, filed in April 1984, three months after the ALJ's decision, Dr. Byrd reported that the claimant had suffered another blackout spell and had been admitted to the hospital for unstable angina. In addition to the four physicians who examined the claimant, two SSA consulting physicians reviewed the claimant’s medical records during the initial application process. On August 3, 1982, Dr. Thurma Jo Fiegel reviewed the records, concluding that the history of congestive heart failure was questionable, the chest pain was not of cardiac origin, and the results of the pulmonary function tests did not meet the levels required by Appendix 1 of the Social Security regulations for disabling chronic pulmonary disease. Another consulting physician, Dr. Patrick Barnes, reviewed the medical records on November 10, 1982 — while claimant’s request for reconsideration of the initial denial was under review — and agreed with Dr. Fiegel’s report with respect to congestive heart failure, chest pain, and lung disease. However, he found the claimant to be moderately impaired and to have a number of job restrictions, including restrictions against working at unprotected heights and driving automotive equipment. During the claimant’s second application process, the SSA consulted another physician, Dr. John R. Adair. Like Dr. Cox before him, Dr. Adair actually examined the claimant. In a report dated July 23, 1983, Dr. Adair confirmed the atrial fibrillation but found the claimant to be “quite suggestible and vague.” R. Vol. II at 163. At the request of Dr. Adair, a medical technician conducted a new set of pulmonary function bests. The technician, while noting the claimant’s complaint that his chest pain was limiting his efforts, concluded that the claimant’s performance represented less than maximum effort. Dr. Adair concluded that he could not detect adequate organic pathology to verify the *1459 extent of the claimant’s subjective physical complaints, citing the “uncertainty of the data base.” R. Vol. II at 165. Finally, another review of the medical records by Dr. Fiegel, in September 1983, produced a cursory and unexplained conclusion that the claimant was slightly impaired. Basically, what we have is two sets of somewhat conflicting medical reports. The treating physicians and Dr. Cox, the first SSA consulting physician, conclude that the claimant has atrial fibrillation and heart disease, accompanied by a number of credited physical symptoms, including dizziness, blackouts, chest pain, shortness of breath and substemal pressure upon exertion. Some of these same physicians also note lung disease, hypertension, and evidence of liver disease. Dr. Byrd characterizes the chest pain and substernal pressure as unstable angina. On the other hand, the remaining three consulting physicians, while noting the evidence of heart and lung problems, do not comment on the liver disease and do not interpret the objective medical evidence as sufficiently supporting the claimant’s extensive physical symptoms. The claimant appeared before the AU without the benefit of counsel. After hearing his testimony and allowing him an opportunity to comment on any perceived errors in the documentary evidence, the AU concluded that the claimant was not disabled as that term is defined by the Social Security Act. The AU first found that the claimant had not been substantially gainfully employed since October 10, 1981, and that he had an impairment. However, he found that the impairment did not meet or equal one of the impairments in Appendix 1 of the Social Security Act regulations, by which the claimant would have automatically qualified as disabled. He therefore went on to explore the claimant’s ability to either return to his past work or to engage in other work, as required by the regulations. The AU found that the claimant could not return to his past work requiring medium to heavy exertion because his residual functional capacity (“RFC”) would not allow it. However, he also found that the claimant’s testimony as to the extent of his impairments was “not wholly credible and ... somewhat probably exaggerated.” R. Vol. II at 17. He found that the claimant retained the RFC for a wide range of light work, restricted only “by inability to work in environments with excessive dust, fumes, or gases.” R.Vol. II at 18. Such a finding recognized no substantial exertional limitations on the claimant’s ability to perform the full range of light work but did explicitly recognize environmental restrictions that are considered nonexertional limitations on the ability to perform that work. The AU, however, went on to conclude that the “claimant’s capacity for the wide range of light [work] has not been significantly compromised by his additional nonexertional limitations.” R. Vol. II at 19. He concluded that Rule 202.14 1 of the Medical-Vocational Guidelines promulgated by the SSA provided the framework for his determination that given the claimant’s residual strength, age, worx experience, and education, he was not disabled. I. Under the Social Security disability insurance provisions, a claimant is disabled “if he is unable to engage in any substantial gainful activity by reason of any medi *1460 cally determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than twelve months.” 42 U.S.C. § 1382c(a)(3)(A) (1982). 2 The Social Security regulations implementing the statutory definition of disability invoke a five-step process for evaluation of a disability claim. If a determination can be made at any of the steps that an individual is or is not disabled, evaluation under a subsequent step is unnecessary. In this case we are concerned with the fifth step. The question is whether the claimant’s impairments, which preclude the performance of past relevant work, also preclude other kinds of substantial gainful work that exist in the national economy. At step five, the Secretary of Health and Humán Services (“Secretary”) bears the burden of presenting evidence that the claimant has the residual functional capacity to do other work that exists in the national economy. Turner v. Heckler, 754 F.2d 326 (10th Cir.1985); Channel v. Heckler, 747 F.2d 577 (10th Cir.1984). The Secretary must consider the claimant’s age, education, and work experience, along with residual functional capacity, in determining whether a claimant can do other work. 20 C.F.R. § 404.1520 (1986). See Tillery v. Schweiker, 713 F.2d 601, 602 (10th Cir.1983). To aid the Secretary with his step-five burden, the SSA has developed Medical-Vocational Guidelines, 20 C.F.R., pt. 404, subpt. P, app. 2 (1986). Sometimes called the “grid” or “grids,” the guidelines relate a claimant’s age, educational background, and work experience to ability to engage in work in the national economy at various levels of exertion (sedentary, light, medium, heavy and very heavy). If the Secretary can determine that an individual claimant precisely fits on the grids, the Secretary can use the grids to decide whether the claimant is able to do other work and is therefore not disabled. The validity of the Medical-Vocational Guidelines was recently upheld in Heckler v. Campbell, 461 U.S. 458, 103 S.Ct. 1952, 76 L.Ed.2d 66 (1983). The Supreme Court sustained use of the grids in appropriate situations as a legitimate method of making disability determinations more uniform. It noted that use of the grids avoided the necessity of calling a vocational expert to testify in each case in which step five of the evaluation process was reached. Instead, the vocational expert could be reserved for special situations where a claimant did not fit the grids and additional evidence was necessary. Use of the Medical-Vocational Guidelines is predicated on an impairment that limits the strength or exertional capacity of the claimant. Therefore, if a claimant’s impairment is of a different nature, the grids may not be fully applicable. For instance, the regulations note that certain mental, sensory, or skin impairments, environmental restrictions, or postural and manipulative restrictions may be independent from exertional limits. 20 C.F.R., pt. 404, subpt. P, app. 2, § 200.00(e) (1986). Where such “nonexertional” limitations combine with exertional limitations which do not in and of themselves establish a disability, then the grids are to provide no more than a framework for determining disability. The hearing officer is not to automatically or mechanically apply the grids but instead must consider all the relevant facts in determining whether the nonexertional limitations diminish the claimant’s ability to perform other work. Id. at § 200.00(e)(2). Similarly, if a claimant’s residual functional capacity does not meet the definition of one of the exertional ranges (sedentary through heavy), then the AU is to “consider the extent of any erosion of the occupational base and assess its significance____ *1461 Where the extent of the erosion of the occupational base is not clear, the adjudicator will need to consult a vocational resource.” Social Security Ruling 83-12. In other words, there are situations where the grids alone cannot yield the answer as to one’s ability to engage in other work in the national economy. II. The primary issue raised on appeal is whether the ALJ had substantial evidence to conclude that the claimant fit the grids and therefore was not disabled. The claimant argues that there was not substantial evidence to support the AU’s finding that the claimant’s nonexertional limitations were insignificant enough to allow application of the grids. Our review of the record reveals an even more fundamental issue, i.e., whether there was substantial evidence to support the AU’s finding that the claimant had the exertional capacity to perform a full range of light work. The first issue is essentially subsumed by the second, since if the claimant lacked exertional capacity to perform light work, there is no need to consider whether nonexertional impairments also imposed limitations on that same capacity. We therefore turn to a review of the claimant’s exertional limitations first. A. Evidence of RFC for a Full Range of Light Work. Implicit in the challenge to the substam tiality of the evidence is the fact that federal court review of the Secretary’s factual findings with respect to disability is limited to a review of whether substantial evidence supports the Secretary’s findings. See 42 U.S.C. § 405(g). The term substantial evidence has been interpreted by the United States Supreme Court to require “more than a mere scintilla. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Richardson v. Perales, 402 U.S. 389, 401, 91 S.Ct. 1420, 1427, 28 L.Ed.2d 842 (1971) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229, 59 S.Ct. 206, 216, 83 L.Ed. 126 (1938)). However, the search for adequate evidence does not allow us to weigh the evidence and substitute our discretion for that of the agency. Cagle v. Califano, 638 F.2d 219 (10th Cir.1981). Nonetheless, we must review the record as a whole, and “the substantiality of the evidence must take into account whatever in the record fairly detracts from its weight.” Universal Camera Corp. v. NLRB, 340 U.S. 474, 488, 71 S.Ct. 456, 464, 95 L.Ed. 456 (1951). The AU appears to view the claimant’s heart, lung, and liver impairments as insignificant limitations on his ability to perform light work because, among other things, he found that the claimant’s testimony was “not wholly credible and ... somewhat probably exaggerated.” R. Vol. II at 17. We acknowledge that it is not our job to make an independent judgment as to the credibility of the claimant’s testimony regarding his subjective physical symptoms. “[T]he determination of credibility is left to the observations made by the Administrative Law Judge as the trier of fact.” Broadbent v. Harris, 698 F.2d 407, 413 (10th Cir.1983). We note, however, that the AU gave no particular reasons for discounting the claimant’s credibility except to observe that his blackouts had occurred when no witnesses were present. Therefore, we feel free to view the AU’s conclusion with “a skeptical eye.” Id. at 414. Additionally, we do not question the AU’s inference that the claimant himself showed that he thought he was able to work by applying for rehabilitation training — training which was denied because of his impairments, using different criteria for disability than the SSA criteria. The AU uses the inference as to the claimant’s own belief, however, as support for the finding that the claimant’s respiratory impairments imposed only an insignificant environmental restriction on his ability to work. While the AU has the discretion to make the initial inference, we note that, logically, the inference does not dictate the conclusion that the claimant thought he could work at a full range of light activity rather than a limited range of light or sedentary activity. The AU attempted to buttress his finding of insignificant environmental restrictions by concluding that the claimant’s combined impairments had not prevented *1462 him from doing light and sedentary work “on a regular and continued basis after October 10, 1981.” R. Vol. II at 17. Light work is defined in the Social Security regulations as work that: involves lifting no more than 20 pounds at a time with frequent lifting or carrying of objects weighing up to 10 pounds. Even though the weight lifted may be very little, a job is in this category when it requires a good deal of walking or standing, or when it involves sitting most of the time with some pushing and pulling of arm or leg controls. To be considered capable of performing a full or wide range of light work, you must have the ability to do substantially all of these activities. 20 C.F.R. § 404.1567(b) (1986) (emphasis added). 3 Sedentary work is defined as work that: involves lifting no more than 10 pounds at a time and occasionally lifting or carrying articles like docket files, ledgers, and small tools. Although a sedentary job is defined as one which involves sitting, a certain amount of walking and standing is often necessary in carrying out job duties. Jobs are sedentary if walking and standing are required occasionally and other sedentary criteria are met. 20 C.F.R. § 404.1567(a) (1986). The ALJ’s conclusion that the claimant had been performing light and sedentary work on a regular and continuing basis after October 10, 1981, is highly questionable. Uncontradicted information in the record as to the claimant’s activities suggests the contrary, and nowhere does the ALJ suggest that this information was discounted by him as unreliable. The record reveals that the claimant tried to do a little painting work and installation of a shower on a contractual basis, as well as to assume a traveling sales job. However, the record also indicates that he was unable to finish or keep these jobs because of his blackouts and strength limitations and that he was never paid for any of these attempts. 4 Also, the record reveals that the claimant’s daily routine was extremely limited. He occasionally drove his wife three miles to the grocery store on back roads, and he tried to walk around his yard every day, but he seldom engaged in any other activity in or outside his home. He could no longer hunt or fish, and even sweeping the floor of his home tired him out, so that he was not able to help his wife with domestic chores. To call such nongainful activity regular and continuous light and sedentary work would seem to defy the record. Other recent decisions by this court support the view that such limited activities in themselves do not establish that one can engage in light or sedentary work activity. In Broadbent v. Harris, 698 F.2d 407 (10th Cir.1983), a claimant had admitted working in his yard, performing a few household tasks, working on cars, and taking occasional car trips. The activities were not conducted on a regular basis, nor did they involve prolonged activity. When considered in the light of the medical testimony, they were judged to be insignificant. While such activities may be considered, along with medical testimony, in determining whether a person is entitled to disability benefits, they do not in themselves establish that a person is able to engage in substantial gainful activity. Markham v. Califano, 601 F.2d 533 (10th Cir.1979). “Ability to drive an automobile, participate *1463 in some community affairs, attend school, or to do some work on an intermittent basis does not necessarily establish that a person is able to engage in a ‘substantial gainful activity,’ but such activities may be considered by the Secretary, along with medical testimony____” Id. at 534. Here, any reasonable reading of the record could not result in use of the claimant’s limited activities to enhance any medical testimony that might support such a finding. The question remains as to whether the medical testimony itself provides substantial evidence that the claimant could engage in a full range of light work. After concluding that the claimant’s liver disease exacerbated the cardiovascular and pulmonary symptoms, the AU found nonetheless that the claimant “can sit, stand, or walk six hours a day and lift and carry 10 pounds frequently and 20 pounds occasionally.” R. Vol. II at 16. While the RFC evaluations made by SSA consulting Doctors Fiegel and Barnes support such a conclusion, they were not explicitly invoked by the AU. Furthermore, neither of these doctors examined the claimant, and their views as to the degree of impairment conflicted with one another. R. Vol. II at 112, 114. For these reasons, their reports have less credence than the contrary views of the treating physicians. See Broadbent v. Harris, 698 F.2d 407, 412 (10th Cir.1983) (opinions of physicians “who have treated a patient over a period of time or who are consulted for purposes of treatment are given greater weight than are reports of physicians employed and paid by the government for the purpose of defending against a disability claim” [quoting Allen v. Califano, 613 F.2d 139 (6th Cir.1980)]). See also Whitney v. Schweiker, 695 F.2d 784, 789 (7th Cir.1982) (treating physician’s report should be favored over that of consulting physician who merely reviews the records, unless treating physician is not credible). The reports of reviewing physicians are also accorded less weight than those of examining physicians. Wier ex rel. Wier v. Heckler, 734 F.2d 955, 963 (3d Cir.1984) (opinions of doctors who have never examined patient have “less probative force, as a general matter, than they would have had if they had treated or examined him”). The conclusion of Dr. Adair, who did examine the patient, also tends to support the AU’s finding, although again it was not specifically invoked by the AU. Also, Dr. Adair only examined the claimant once, and the treating physicians’ abilities to observe the claimant over a longer period of time should be given due weight when compared to speculation by a consulting physician regarding the validity of claimant’s symptoms. Cf. Allen v. Weinberger, 552 F.2d 781, 786 (7th Cir.1977) (medical opinion that claimant might be able to perform light work was given “little weight,” since inherently speculative and physician examined patient only once). In this case, the AU, rather than explicitly accepting the views of Doctors Adair, Fiegel, and Barnes and rejecting the views of the treating physicians and Dr. Cox, instead attempts to bolster his finding by invoking the December 12, 1983, RFC evaluation of Dr. Byrd, one of the treating physicians. The AU states that Dr. Byrd’s evaluation “was generally consistent with light work which allowed alternate sitting, standing, and walking.” R. Vol. II at 16. Such a statement both mischaracterizes the evaluation and assumes that alternately sitting, standing, and walking allows one to perform a full range of light work. What Dr. Byrd’s evaluation concluded was that (1) the claimant could neither sit, walk, nor stand for longer than two hours at a time for a maximum of a six-hour work day. R. Vol. II at 199. Alternate sitting, standing, or walking by implication precludes the kind of extensive sitting, standing and walking contemplated by the definition of light activity. To elaborate, being able to sit, stand, or walk alternately for only six hours collectively would seem to impose significant restrictions on ability to perform light work, since light work by definition is work that “requires a good deal of walking or standing, or ... involves sitting most of the time with some pushing and pulling of arm or leg controls.” 20 C.F.R. § 404.1567(b). 5 It *1464 is by no means obvious that being able to walk and stand collectively for only half an eight-hour day and to sit for only two hours at a time would allow the claimant to perform substantially all of the jobs in the range of light work. Cf Rivers v. Heckler, 577 F.Supp. 766 (S.D.N.Y.1984) (where durational capacity is no more than two hours a day for sitting, standing or walking, the claimant is not capable of light work). Furthermore, Dr. Byrd limited the claimant’s combined activities to six hours out of an eight-hour day, i.e., to three-fourths of a full day’s work. Being able to work less than full time would also seem to preclude the ability to perform substantially all of the jobs in the range of light work. As the court noted in McCoy v. Schweiker, 683 F.2d 1138 (8th Cir.1982): “[t]he RFC that must be found if the grid is to be used, in the case of ... light work, is ... the ability to perform the requisite physical acts day in and day out, in the sometimes competitive and stressful conditions in which real people work in the real world.” Id. at 1147 (footnote omitted). The ALJ’s invocation of Dr. Byrd’s report in apparent support of the decision that the claimant could perform most jobs in the light work category will not substitute for the failure to explain what was in actuality the rejection of the views of the treating physicians and Dr. Cox. If the medical opinions of treating physicians are to be rejected, specific, legitimate reasons for so doing must be set forth. Turner v. Heckler, 754 F.2d 326 (10th Cir.1985); Byron v. Heckler, 742 F.2d 1232 (10th Cir.1984); Murray v. Heckler, 722 F.2d 499 (9th Cir.1983). Here the AU avoided any explanation by attempting to ignore or downplay the conflicting medical opinions. In summary, rather than evaluating and rejecting the reports of the treating physicians and Dr. Cox, which essentially confirm one another, are based on objective medical evidence, and reflect observation over time, the AU instead mischaracterizes Dr. Byrd’s evaluation. Instead of challenging the claimant’s reports of his activities, the AU mischaracterizes their nature. What is left to support the decision that the claimant can perform light work is the AU’s impression that the claimant exaggerated his physical symptoms. The decision is hanging on this single thread of evidence. Such a thread seems to us to constitute a mere scintilla. Our review of the record as a whole shows an absence of substantial evidence that the claimant has the exertional capacity to perform a wide range of light work. B. Significance of Nonexertional Limitations. Although our conclusion regarding the claimant’s RFC negates the need to address the claimant’s nonexertional limitations, we choose to comment briefly on this aspect of the AU’s decision. First, we note that there is substantial evidence in the record to suggest that the claimant needed to restrict the environment in which he performed light work. Dr. Wallace Byrd, one of the treating, physicians, concluded that the claimant should not work where dust, gases, or fumes were present. Additionally, both Doctors Cox and Matter concluded that the claimant had a chronic obstructive lung condition. . Finally, the AU gave the claimant “the benefit of the doubt” and credited the claimant’s testimony of his fatigability. R. Vol. II at 16. Therefore, and in spite of the somewhat inconsistent results of studies of the claimant’s impaired pulmonary function, the AU concluded that the claimant “probably has a mild to moderate obstructive lung condition and should avoid work in environments with excessive dust, fumes, or gases.” R. Yol. II at 16. In contrast, although there is substantial evidence that the claimant needed to restrict the environment in which he performed light work, there is little evi *1465 dence to support the ALJ’s finding that the environmental restriction placed an insignificant limitation on the claimant’s ability to perform substantially all of the jobs in the light work classification. Admittedly, a nonexertional impairment can have a negligible effect on the range of jobs available. 43 Fed.Reg. at 55,358 (1978). See Channel v. Heckler, 747 F.2d 577, 582 n. 6 (10th Cir.1984). Cf. Grant v. Schweiker, 699 F.2d 189, 192 (4th Cir.1983) (use of grids may not be relied upon to the extent that nonexertional impairments further limit the range of jobs available). The AU, however, must back such a finding of negligible effect with the evidence to substantiate it. Cf. Odle v. Heckler, 707 F.2d 439 (9th Cir.1983) (substantial evidence supported finding that nonexertional impairments did not significantly limit claimant’s exertional capabilities); Cummins v. Schweiker, 670 F.2d 81, 84 (7th Cir.1982) (substantial evidence supported AU’s finding that claimant’s visual impairment would not interfere with performance of sedentary work and grids were therefore applicable); Kirk v. Secretary of Health & Human Servs., 667 F.2d 524, 536-37 (6th Cir.1981) (substantial evidence supported AU’s determination that claimant’s mental impairments did not significantly limit the range of work permitted by his exertional capacity), cert. denied, 461 U.S. 957, 103 S.Ct. 2428, 77 L.Ed.2d 1315 (1983). The Secretary argues that the AU’s use of the term “excessive” modifying dust, fumes, and gases in itself establishes that the restriction is insignificant. However, such an argument is straining at gnats and swallowing camels. Nowhere does the AU indicate what constitutes “excessive” dust, fumes, or gases. Additionally, a mere adjective does not substitute for careful explanation of the weight given to conflicting evidence — an explanation that, as we have previously noted, is missing in this case. If the record had supported the AU’s finding that claimant’s exertional capacities allowed him to perform a full range of light work, we would have been obligated to address the claimant’s contention that a remand for vocational testimony was required. Arguably, only vocational testimony could have provided sufficient data as to whether substantially all of the jobs in the light work category could accommodate the claimant’s environmental restrictions. 6 Given our conclusion here, we need not resolve this question. And although vocational testimony might show that the environmental limitations were insignificant, such testimony could not overcome the substantial evidence that the claimant’s exertional limitations took him out of the light work category. CONCLUSION Because we find insubstantial support in the record for the AU’s finding that the claimant could perform a full range of light work and substantial evidence that he could not, we reverse. We take due note of the fact that if the AU had found the claimant incapable of light, medium, or heavy exertion, then the grids would have dictated a finding of disability, given the claimant’s advancing age, the nontransfer *1466 ability of his job skills to sedentary work, and his educational level. Where the burden is on the Secretary at step five of the disability process to produce evidence that the claimant can perform other work, and the Secretary does not meet that burden, reversal is appropriate. Broadbent v. Hams, 698 F.2d 407, 414 (10th Cir.1983) (reversal proper where prima facie case not sufficiently rebutted by Secretary and where Secretary’s decision not supported by substantial evidence). Because we hold for the claimant, we need not address his additional claim that his lack of counsel deprived him of a fair hearing. We hold that the claimant is entitled to disability benefits and remand for such an award from the date of his second application for Social Security benefits, by which time substantial evidence of his disability was in the record. Reversed. Footnotes [1] [[Image here]] 20 C.F.R., pt. 404, subpt. P, app. 2, § 202.00, Table No. 2, Rule 202.14. [2] The statute also states: [A]n individual shall be determined to be under a disability only if his physical or mental impairment or impairments are of such severity that he is not only unable to do his previous work but cannot, considering his age, education, and work experience, engage in any other kind of substantial gainful work which exists in the national economy, regardless of whether such work exists in the immediate area in which he lives, or whether a specific job vacancy exists for him, or whether he would be hired if he applied for work. 42 U.S.C. § 1382c(a)(3)(B) (1982). [3] The report accompanying the promulgation of the Medical-Vocational Guidelines elaborates on the exertional categories by stating that “within a range of work (sedentary, light, medium, heavy or very heavy) unless the individual possesses physical capacities equal to the strength requirements for most of the jobs in that range, he or she cannot be classified as able to do the pertinent range of work." 43 Fed.Reg. at 55,361 (1978) (emphasis original). In other words, while a claimant need not be able to perform all jobs in the range in order to fall within the range, one must be able to perform a substantial majority of them. See abo O’Leary v. Schweiker, 710 F.2d 1334, 1339 (8th Cir.1983): Santise v. Schweiker, 676 F.2d 925, 934 (3d Cir.1982); Kirk v. Secretary of Heatth & Human Servs., 667 F.2d 524, 537 (6th Cir.1981), cert. denied, 461 U.S. 957, 103 S.Ct. 2428, 77 L.Ed.2d 1315 (1983); Davis v. Schweiker, 536 F.Supp. 90, 100 (N.D.Cal.1982). [4] The record does not mention whether these activities would be considered light work, but if they required lifting twenty pounds or less, along with a good deal of walking or standing or prolonged sitting, they would qualify. [5] Social Security Ruling 83-12, dealing with the medical-vocational rules in situations where they do not precisely match a claimant’s exer *1464 tional abilities, specifically mentions the special situation in which a claimant can only sit or stand alternately. The ruling states that where an individual must alternate sitting or standing, "[s]uch an individual is not functionally capable of doing either the prolonged sitting contemplated in the definitions of sedentary work (and for the relatively few light jobs which are performed primarily in a seated position) or the prolonged standing or walking contemplated for most light work.” [6] A remand for some kind of additional proceedings would be appropriate where an ALT made no findings as to nonexertional impairments or where minimal findings were not supported by adequate evaluation of the evidence in the record. See Channel v. Heckler, 747 F.2d 577 (remanded for findings as to whether claimant’s nonexertional skin impairments precluded performance of full range of sedentary jobs on a sustained basis). Also, where the Secretary failed to meet his burden by failing to introduce vocational testimony that was material to the outcome, vocational testimony has been ordered. See Allen v. Secretary of Health & Human Servs., 726 F.2d 1470 (9th Cir.1984) (remanded for vocational testimony as to whether claimant's environmental restrictions negated claimant’s ability to perform a full range of light work); Grant v. Schweiker, 699 F.2d 189 (4th Cir.1983) (remanded for vocational testimony where claimant demonstrated nonexertional impairments limiting the range of jobs available to claimant). Some kinds of nonexertional impairments may lend themselves to a determination by the ALT as to whether their impact on the range of jobs is significant. See Channel, 747 F.2d at 582. We decline, however, to determine whether environmental restrictions are by their nature the kind of nonexertional impairment that requires vocational testimony in order to determine the extent of the erosion of the occupational base.