On April 23,1990, the Board of Veterans’ Appeals (BVA) rendered a decision denying the appellant, Thomas C. Connolly, service connection for his alleged Meniere’s disease. Upon consideration of the record and briefs of the parties, the Court holds that the Department of Veterans Affairs (VA) violated its duty to assist, and that the BVA’s decision failed to state the medical evidence it relied upon in reaching its medical conclusions. Accordingly, we vacate and remand the decision of the BVA.
I.
Connolly joined the United States Navy on September 25, 1944. On July 10, 1945, he went to sickbay complaining of pain and discharge in his left ear, headaches, and hearing loss. Connolly’s left eardrum was found to be red, thickened, and distorted, with a perforation in the posterior half. The diagnosis at that time was acute otitis media (inflammation of the middle ear). Connolly was treated in sickbay for the next five weeks with sulfonamides and penicillin, but showed little improvement in his condition. He was then transferred to the Naval Hospital in Shoemaker, California.
An examination on August 20, 1945, found that the perforation on Connolly’s left eardrum was healing, and that the discharge from that ear was diminishing. The clinical impression recorded at this time was chronic otitis media. Connolly was continued on penicillin and sulfonam-ide treatment, and his condition steadily *568improved. On September 19, 1945, he was discharged from the hospital and returned to duty. On November 20, 1945, Connolly was discharged from the Navy for an unrelated, non-service-connected disability.
After filing a disability compensation claim with the VA for alleged ear difficulties, Connolly underwent a VA examination on November 21, 1958. At that time Connolly reported having had “stopped up” ears in rainy weather and occasional dizzy spells since his discharge from the service. He also reported that the discharge from his ears had not ceased until 1950. The VA examination found, among other things, that Connolly had slight scarring on both his eardrums, and that he was suffering from some hearing loss in both ears. As a result of this examination, Connolly was granted service connection for “otitis media, left, non-suppurative,” but given a non-compensable rating, effective October 11, 1957.
In a March 6,1978, letter from the VA to Connolly, it was noted that effective March 15,1976, Connolly’s disability rating for his otitis media had been increased to 20% as the result of VA medical examinations which had shown that Connolly’s hearing loss had increased. The letter also noted that both the Pittsburgh, Pennsylvania, Regional Office (Regional Office) and the BVA had recently turned down Connolly’s requests for a separate service connection for his tinnitus, dizziness, and nausea, on the grounds that those conditions were merely symptoms of Connolly’s already service-connected otitis media.
In May 1988, Connolly requested an increase in his disability rating for his ear disorders. After the Regional Office denied this claim, Connolly appealed to the BVA, raising the possibility that he might be suffering from service-connected Meni-ere’s disease.
I had problems for a few years [after service] and in 1954 after consulting with our family physician he informed me; loss of hearing, tinnitus, loss of balance and upset stomach add up to one thing; you have Meniere’s Disease....
R. at 76. (Connolly had previously implied that he might be suffering from Meniere’s disease in a 1976 letter to the VA, stating that at some unspecified time in the past a family doctor had given him a “diagnosis of Meniere’s.” R. at 55.) The VA then requested an examination to determine the “current nature and extent,” R. at 86, of Connolly’s disabilities.
In response to this VA request, Connolly underwent outpatient treatment and examination at a local VA medical facility during January and February of 1989. During this period, Connolly stated that he was suffering from nausea, dizziness, tinnitus, headaches, and hearing loss. He also stated that the severity of his symptoms, though generally progressively increasing, would wax and wane in an episodic manner. A laboratory report from the Eye and Ear Hospital of Pittsburgh found that Connolly was suffering from balance difficulties and reduced vestibular functioning bilaterally, and the treating VA doctor noted that Connolly was suffering from bilateral hearing loss, and that his symptoms were “possibly consistent with bilateral Meni-ere’s.” R. at 89 (emphasis in original). Connolly, however, stated on numerous occasions that he “wanted no further testing,” R. at 89-90, and eventually the treating doctor terminated the examination and treatment, in accordance with Connolly’s desires, without reaching a final diagnosis.
The Regional Office denied Connolly’s claim for service connection for Meniere’s disease on June 6, 1989. Connolly then appealed to the BVA, repeating his allegations regarding his symptoms and asserting that he had been suffering from these symptoms since his in-service ear infection. On April 23, 1990, the BVA denied Connolly’s claim for service connection for Meni-ere’s disease.
In the present case, service medical records are entirely negative for evidence of Meniere’s disease. The earliest clinical indication of the possible presence of Meniere’s disease is revealed by the VA examination conducted during the months of January and February 1989, approximately 44 years following service discharge, at which time there *569was noted the presence of reduced vestibular function, bilaterally. Following a thorough review of the evidence, with the application of generally accepted medical principles, we are unable to reasonably associate the veteran’s current Meniere-like symptoms, first persuasively documented many years following service discharge, with any incident of his period of his active military service. Nor has it been persuasively demonstrated that such symptomatology is in any way causally related to the veteran’s service-connected otitis media and accompanying hearing loss. Accordingly, we are unable to reach a favorable decision in this case.
Thomas C. Connolly, BVA 90-14474 at 3-4 (April 23, 1990). Connolly subsequently perfected an appeal to this Court. .
II.
At least since its March 6, 1978, letter to Connolly, the VA has acknowledged that Connolly suffers from dizziness, sensory hearing loss, nausea, and tinnitus. Moreover, the VA doctor who was examining Connolly in 1989 to determine the “current nature” of his disability, indicated that as a result of these symptoms, Connolly may well be suffering from bilateral Meniere’s disease. Therefore, it is the opinion of this Court that Connolly had a well-grounded claim for service connection for Meniere’s disease, and thus the VA had the duty to assist him in developing this claim. See 38 U.S.C. § 5107(a) (formerly § 3007(a)), Murphy v. Derwinski, 1 Vet.App. 78 (1990).
The VA’s duty to assist arises out of its long tradition of ex parte proceedings and paternalism toward the veteran. The duty to assist has been codified, 38 U.S.C. § 5107(a), and is explicitly recognized by the VA in its own regulations, 38 C.F.R. § 3.103(a) (1991). Among the actions the duty to assist has been found to require from the VA are, when appropriate, a duty to “conduct a thorough and contemporaneous medical examination_”, EF v. Derwinski, 1 Vet.App. 324, 326 (1991) (quoting Green v. Derwinski, 1 Vet.App. 121, 124 (1991)); a duty to put a veteran “adequately on notice that more [is] required from him”, Wood v. Derwinski, 1 Vet.App. 190, 193 (1991); and a duty to be “helpful and clear in explaining to a veteran what evidence [the veteran] needs together with help and advice in obtaining it.” Wood, at 192-93.
It would have clearly assisted Connolly in his claim if he had had a definite diagnosis of whether or not he was suffering from Meniere’s disease. The VA, however, did not make an adequate effort to obtain such a diagnosis. While it is true that Connolly had on a number of occasions expressed a desire that no more tests be conducted upon him, the record does not reflect that the VA properly explained to him the importance of a definite diagnosis. See Wood, at 192-93. If Connolly had decided after receiving this information that he did not wish to have any further examinations, then there would have been nothing more the VA could do. However, in failing to give Connolly the opportunity to make an informed decision regarding whether to undergo additional examination, the VA violated its duty to assist.
III.
In coming to its ultimate conclusion that Connolly’s “Meniere-like” symptoms were unrelated to his in-service ear infection, the BVA relied on unstated “generally accepted medical principles,” Connolly, BVA 90-14474 at 5. The BVA’s failure to explicitly name and discuss these “generally accepted medical principles,” however, is a violation of the principle expressed in Murphy v. Derwinski, 1 Vet.App. 78 (1990) and Colvin v. Derwinski, 1 Vet.App. 171 (1991).
In Murphy, the Court stated that:
the fact that a BVA panel may include a physician is not by itself sufficient for the purposes of [38 U.S.C. § 7104(d)(1) (formerly § 4004(d)(1))]_ BVA decisions must include the ‘reasons or bases’ for medical conclusions, even those opined by the BVA physician; a mere statement of an opinion, without more, does not provide an opportunity ... for *570this Court to review the BVA decision ‘on the record’ as required by 38 U.S.C. § [7252(b) (formerly § 4052(b)].
Murphy, at 81. In Colvin, the Court expounded further on this principle, stating,
BVA panels may consider only independent medical evidence to support their findings. If the medical evidence of record is insufficient, or in the opinion of the BVA, of doubtful weight or creditability, the BVA is always free to supplement the record by seeking advisory opinions, ordering a medical examination or citing recognized medical treatises in its decisions to support its ultimate conclusions.
Colvin, at 174-75.
Therefore, on remand the BVA must explicitly state and discuss the medical evidence it relies upon in determining whether Connolly actually suffers from Meniere’s disease, and if so, what is the relationship between that disease and Connolly’s in-service ear condition.
IV.
For the foregoing reasons, the decision of the BVA is VACATED and REMANDED for further proceedings consistent with this opinion.