On January 15, 2000, Charles P. Galatis, then seventy-six years old, was admitted to Massachusetts General Hospital (MGH). Once admitted, he was diagnosed with stage IV lung cancer, and over the ensuing weeks he suffered a rapid over-all decline in his physical and mental condition. Galatis remained hospitalized,1 and he died on February 25, 2000.
On February 9, 2000, Galatis executed a document purporting to be his will. The executor named in the will formally presented it for probate, joined by the will’s principal beneficiary, the town of Skiathos, Greece.2 Two of the decedent’s cousins contested the *274will. After a ten-day trial, a Probate and Family Court judge declined to allow the will, because she found that Galatis lacked testamentary capacity on February 9. Because that finding is amply supported by record evidence, we affirm.
Background. The judge made 559 factual findings that totaled seventy-one pages. We summarize those findings, almost all of which are unchallenged, and highlight the facts still in dispute. See Rempelakis v. Russell, 65 Mass. App. Ct. 557, 559 (2006).
1. Galatis’s background medical conditions. By the time Galatis was admitted to MGH, he already suffered from a long list of medical problems including diabetes, hyperkalemia (excess potassium in the blood), and major depression. For such problems, Galatis was taking twelve different prescribed medications, including the antidepressant Elavil, and two different narcotics for pain relief. The symptoms of anxiety and depression worsened following his diagnosis with metastatic lung cancer. He therefore was prescribed a second antidepressant, and the dosages for both antidepressants subsequently were increased. His painkillers also were aggressively increased, and he was placed on a self-administered morphine drip beginning on February 8.
2. The February 1 document. When he entered MGH, Galatis apparently had no existing will. At some point during his initial hospitalization, he asked a friend, Steven Damaskos, to record his thoughts regarding the disposition of his property. Damaskos wrote down individual names that Galatis provided to him, followed by specific amounts of money they were to receive. Damaskos then transcribed the information onto a will template form he found through an Internet search, and Galatis signed the resulting document on February 1, 2000. As discussed below, although the February 1 document was admitted in evidence at the trial on the February 9 will, it was not itself offered for probate prior to the conclusion of that trial.
3. Differences between the February 1 document and the February 9 will. An attorney brought in by Damaskos to prepare a proper will for Galatis was given the February 1 document as a starting point. The end product was the will executed by Galatis on February 9, 2000. The estate plan delineated in that will is similar in most material respects to that contained in the earlier *275document. Both documents provided for the creation of a charitable -trust administered by a foundation in Skiathos, the corpus of which\would consist of Galatis’s real estate assets in Greece. Unsurprisingly, the later document drafted by the attorney generally is a more developed product than the earlier one produced by Damaskos, and many of the differences could best be described as refinements or minor modifications to the earlier plan. For example, the February 9 will provided for the disposition of certain personal property not enumerated in the February 1 document. However, the February 9 will also included some curious additions and omissions. The will included one beneficiary who — according to uncontradicted testimony — apparently does not exist, and the residuary clause included in the February 1 document was omitted (even though — again, according to uncontra-dicted testimony — the will did not otherwise dispose of all of Galatis’s estate).3
4. The events of February 8, 2000. Sometime in the late afternoon of February 8, 2000 (the day before the will signing), Galatis was administered a dose of Ativan for his anxiety symptoms. Apparently as a result of an adverse reaction to the Ativan, Galatis became somnolent and developed a facial droop. He exhibited slurred speech, drowsiness, and inability to pay attention, and he required constant stimulation to generate answers to questions. After it was observed that his arms were flopping, he was diagnosed with asterixis, which is indicative of central nervous system impairment. These symptoms also led to Galatis being diagnosed with the underlying brain condition encephalopathy.
5. The extent of Galatis’s temporary recovery. On the evening of February 8, Galatis was given a drug to try to reverse the effects of the Ativan. Both sides agree that the administration of this drug (the antidote) had some beneficial effects. They differ sharply, however, on the import of the Ativan episode with regard to Galatis’s mental state when he signed the will the following day. The will proponents sought to portray the administration of the antidote as allowing Galatis to make a dramatic recovery *276through which he regained sufficient capacity to execute the will. The will contestants asserted that while the reaction to the Ativan may have exacerbated Galatis’s mental condition on February 8, he by that time was already suffering from encephalopathy, which was impairing his cognitive and motor functions.
6. The will contestant witnesses. The will contestants supported their case with expert opinion testimony from Dr. Marc Whaley, a forensic psychiatrist. Dr. Whaley testified from his review of the medical records that Galatis had exhibited symptoms of encephalopathy (albeit less severe) in the days prior to the Ativan reaction.4 He also opined that even though Galatis may have shown some improvement through the administration of the antidote, the underlying encephalopathy would not have abated. According to him, a person with encephalopathy suffers from a loss of ability to communicate, remain oriented, think logically, solve problems, and remember information.
The will contestants also called Dr. John Stoeckle, the attending physician who directed Galatis’s care at MGH. Dr. Stoeckle was particularly well-suited to offer an opinion regarding Galatis’s mental capacity because he had served as Galatis’s primary care physician for the last fifteen years of his life. In addition, Dr. Stoeckle had examined Galatis throughout his hospitalization, including multiple times each day on February 6 through 10. Consistent with the testimony of Dr. Whaley, Dr. Stoeckle opined that Galatis suffered from encephalopathy and that as of February 9, this condition impaired his ability to think clearly, orient himself, speak and communicate, think logically, solve problems, and remember information.
7. The contemporaneous hospital records. The hospital records from February 9 (the day Galatis signed the will) provided some fodder for each side. A nurse’s progress note from 6:00 a.m. on February 9 indicates that Galatis had suffered from the effects of Ativan throughout the night and continued to be “weak” and “confused [as] to place.” One of the treating physicians stated in a note from the early afternoon that Galatis had “recovered from last night’s events” and “seem[ed] back to baseline.” However, a nurse’s note indicated that as of 1:00 p.m., although Galatis was oriented to himself, his whereabouts, and the time, he complained *277of “general confusion.” In addition, a social worker who met with Galatis at 2:00 p.m. on February 9 noted that he was “[difficult to interview because of mental status changes,” and a nursing note at 7:00 p.m. stated that, although Galatis’s condition had improved from the previous day, he continued to be “sleepy” and “confused” at times. Assessing Galatis’s mental state when the will was signed on February 9 against the backdrop of the medical records was made particularly challenging by the fact that no testimony or other evidence established at what time that day the signing actually took place.5
8. The will proponents’ witnesses. The will proponents offered the testimony of the attorney who drafted the February 9 will. Although she attended the will signing, she had no personal recollection of when it was signed, nor could she produce any contemporaneous notes of the event. She did testify that, in her opinion, Galatis was of “sound mind” at the time he signed the will, but the judge discounted such testimony due to the fact that she lacked any records of ever meeting with Galatis alone at any point prior to the will execution, her “defensive demeanor” while testifying, and the fact that her claim to have met alone with Galatis was contradicted by that of Damaskos (Galatis’s friend and the author of the February 1 document). In fact, the judge sharply criticized the lawyer for failing to inquire at all, at any point prior to the will execution, as to Galatis’s mental status or medical diagnosis.
The will proponents also presented the testimony of the two nurses who attested to the will signing. One had virtually no personal recollection of the will signing, and she admitted that her opinion was based on her general practice not to witness a will unless the testator was of sound mind. The other attesting nurse recalled that Galatis seemed “alert” at the time of the will signing.6 In addition, the proponents presented the testimony of Dr. Ernie Paul Barrette, a physician who treated Galatis at MGH. *278Although Dr. Barrette expressed his opinion that Galatis had testamentary capacity on February 9, such testimony was undercut by several factors: even though Dr. Barrette was testifying only as a treating physician, not as an expert on testamentary capacity, he had no personal recollection of treating Galatis, and in any event he stopped treating Galatis on February 2.7
There was also testimony from several friends and family members, many of whom visited Galatis while he was hospitalized. Of the individuals who came to the hospital, only two were present at some point on the day of the will signing, Damaskos and Demetrios Skopas. Damaskos, although not a blood relative of Galatis, considered him to be like an uncle. Damaskos visited Galatis at the hospital three or four times a week, and testified in general terms that he found Galatis to be “clear minded” and “able to communicate.” Damaskos was not present at the will signing itself. Skopas, ninety-one years of age at the time of trial, was a fixture at Galatis’s bedside, spending every evening at the hospital and staying overnight approximately fifteen times. He testified that “he had no difficulty communicating with” Galatis until the day of his death. However, Skopas also testified that he remained in Galatis’s hospital room during the will signing, and this testimony revealed that he may have had some concern about Galatis’s physical and mental condition.8
Discussion. 1. Galatis’s testamentary capacity on February 9, 2000. As noted, the will proponents formally presented the February 9 will for probate, and thus the trial focused on whether Galatis had testamentary capacity on the date that will was executed.9 Whether a testator had testamentary capacity is a question of fact. Duchesneau v. Jaskoviak, 360 Mass. 730, 733 *279(1972). On appeal, “[i]t is our obligation to review the evidence and reach a decision in accordance with our own reasoning and understanding, giving due weight to the findings of the trial judge, which we will not reverse unless they are plainly wrong.” Paine v. Sullivan, 79 Mass. App. Ct. 811, 811-812 (2011), quoting from Palmer v. Palmer, 23 Mass. App. Ct. 245, 249-250 (1986). Although there is a presumption that the testator had testamentary capacity, once the contestants produce “some evidence of lack of testamentary capacity, the presumption of [capacity] loses effect” and the burden shifts to the proponents to prove by a preponderance of the evidence that the testator was able
“to understand and carry in mind, in a general way, the nature and situation of his property and his relations to those persons who would naturally have some claim to his remembrance[,] . . . freedom from delusion which is the effect of disease or weakness and which might influence the disposition of his property[,] [a]nd . . . ability at the time of execution ... to comprehend the nature of the act of making a will.”
Palmer v. Palmer, 23 Mass. App. Ct. at 250, quoting from Goddard v. Dupree, 322 Mass. 247, 250 (1948).
We agree with the trial judge that the evidence of incapacity summarized above was sufficient to shift the burden of proof to the will proponents, and we discern no clear error in the judge’s finding that they failed to carry that burden. There is no merit to the proponents’ contention that the trial judge’s finding of incapacity was based on evidence of only “general illness and depression.” To the contrary, both Dr. Stoeckle (Galatis’s long-term physician) and Dr. Whaley specifically testified that Galatis’s diagnosed encephalopathy and myriad medications prevented him from being able to read or understand the provisions of a will on February 9, 2000. That view also found support in other evidence, such as the contemporaneous hospital records and the marked deterioration in the legibility of Galatis’s signature on the will (indicative of significantly impaired motor function associated with encephalopathy). To be sure, there was some evidence to support the will proponents’ position that Galatis had regained testamentary capacity on February 9 (at whatever time that day he executed the will), but it was the judge’s role as fact finder to *280assess all the evidence and to resolve any conflicts.10 The record reveals that she carried out her fact-finding duties with uncommon care and comprehensiveness. In sum, the judge’s finding that Galatis lacked testamentary capacity to execute the February 9 will enjoys ample support in the trial evidence.
2. Status of the February 1 document. Finally, we are called upon to address a procedural loose end regarding the February 1, 2000, document. Neither will proponent petitioned to have that document probated as Galatis’s will in the event the February 9 will were disallowed. Hence the trial judge ruled that whether the February 1 document could be considered a valid will was not before her.11 Nevertheless, she specifically found that the contestants to the February 9 will did not submit sufficient evidence to overcome the presumption that Galatis had capacity on February 1, 2000, thus raising the possibility that the document that he signed that day could qualify as his will. The will proponents argue that the judge erred in not ruling whether the February 1 document constituted a valid will. Their argument rests on the *281premise that a particular procedural step that one of them took sufficed to present the February 1 document for probate. Assessing the validity of that premise requires additional background detail.
The record shows that, early in the litigation, the proponents of the February 9 will had some confusion about what they should do with the February 1 document. On May 21, 2007, Skiathos filed a “Motion for Instructions” concerning that issue. In its motion, Skiathos expressed its lack of clarity as follows:
“It is unclear to the parties whether the Court would like a copy of the February 1, 2000 Will filed with this matter, if the Court would like a petition filed for the February 1, 2000 Will or whether the Court would want a Motion to Allow this Will in the Alternative filed in this matter.”
A close reading of this awkwardly phrased sentence reveals that Skiathos was seeking guidance on two distinct issues: first, whether it should file a formal petition to have the February 1 document presented for probate, and second, whether it could file a copy of that document in lieu of the original because the original could not be located. A judge different from the trial judge endorsed the motion as “allowed,” together with a notation that read, “[A]s no original exists, filing will not be required.” In other words, the judge expressly addressed only the issue regarding the missing original, not what procedure Skiathos should use if it wanted to present the February 1 document for probate.
We acknowledge that the motion judge’s response to the motion for instructions may not have alleviated Skiathos’s confusion about what procedure it should have followed if it had wanted to present the February 1 document for probate. However, precisely because of that remaining uncertainty, it was incumbent on the proponents to seek further clarification from the court. See Coyne Industrial Laundry of Schenectady, Inc. v. Gould, 359 Mass. 269, 275-276 (1971). Neither the will proponents nor the executor did so. In any event, the judge’s incomplete response to the awkwardly phrased motion did not relieve the will proponents of their obligation to follow proper procedures. See Ferriter v. Borthwick, 346 Mass. 391, 393 (1963) (“In situations where there is more than one will it is within the power of the judge to require petitions to be filed to probate each will and to hear them together”). Because the February 1 document was not presented *282for probate,12 the trial judge correctly concluded that whether it should have been allowed as Galatis’s will was not properly before her.13
Judgment affirmed.
On February 2, 2000, Galatis was transferred from MGH to, Spaulding Rehabilitation Hospital (Spaulding). On February 6, he was readmitted to MGH after developing renal failure and a urinary tract infection.
The decedent was born in Skiathos in 1923. He was an only child and had no children, and his wife of thirty-three years died from lung cancer in 1995 or *2741996. He was survived by his cousins, George Kyparissos, Charalambos Kyparissos, and Athena Perisiadou. During the pendency of the probate of Galatis’s estate, Charalambos Kyparissos died and his heirs-at-law were substituted for him.
Additionally, although both documents provided that Skiathos would be the beneficiary of a trust funded by Galatis’s real estate holdings there, the February 9 will added that his home in Skiathos be converted into a museum and never be sold. At the same time, the February 9 will — unlike the February 1 document — would grant the executor a power of sale of real property, thus obviating the need for him or her to seek court approval before selling the Skiathos house. See Samra v. Yuan, 40 Mass. App. Ct. 934, 935-936 (1996).
Dr. Whaley testified that the encephalopathy was likely caused by a combination of factors, including the effect of the interactions of the numerous drugs he was taking and electrolyte imbalance from his infection, as well as multiple derangement due to renal failure.
On appeal, the will proponents argue that it can be inferred that the will was likely signed sometime between 2:30 p.m. and 7:00 p.m. A note recorded by Dr. Stoeckle indicates that Galatis was to receive radiation therapy at 2:30 p.m., and a nurse’s note from 7:00 p.m. says, “Friends here all day. [Galatis] [d]oes not want to be left alone. Family members going over [patient’s] legal paper [and] will today.”
Dr. Whaley and Dr. Stoeckle both explained in their testimony that the term “alert” is a medical term of art meaning only that a patient is conscious and aware of his surroundings, and that the term does not refer to a patient’s mental capacity.
The judge did not abuse her discretion in ruling that Dr. Barrette did not qualify as an expert on testamentary capacity. Nor did the judge abuse her discretion in precluding the will proponents from calling a different witness as an expert based on discovery violations. As the will proponents accurately point out, even though Dr. Barrette was never qualified as an expert on this, he did — in response to a question asked on cross-examination — state his opinion on Galatis’s mental capacity on February 9 based on his review of the medical records. However, the admission of such opinion evidence of course does not mean that the judge was required to credit it.
During the signing, Skopas asked Galatis to write his signature more clearly, to which Galatis responded, “Leave me alone.” After the signing, according to Skopas, Galatis asked him, “I did everything right, right?” Skopas also testified that Galatis seemed “upset” ai)d that “his hand was shaking.”
The will contestants also alleged undue influence. The judge rejected that claim and the correctness of her findings and rulings on that point are not before us.
The proponents suggest that the judge was required to reject Dr. Stoeckle’s testimony regarding his patient’s mental capacity because, in response to a question on cross-examination about whether he understood the term “testamentary capacity,” Dr. Stoeckle said, “I think I do not because I don’t think I’ve ever had anyone ask me to do it and I don’t remember discussing it with any colleague or lawyer or anybody.” The force of this admission, however, is greatly diminished when it is viewed in the context of the rest of Dr. Stoeckle’s testimony. Immediately after acknowledging his lack of familiarity with the legal term “testamentary capacity,” Dr. Stoeckle clarified that he understood the concept “in a conceptual way, but not in a[n] operational way.” On redirect, Dr. Stoeckle again clarified that, irrespective of the legal terminology, it was his view that Galatis suffered from “mental impairment and incapacity in comprehending, appreciating, understanding the nature, significance and consequences of the contents and execution of a will.” In any event, the judge specifically limited Dr. Stoeckle’s testimony to his opinion as Galatis’s treating physician. Accordingly, in her findings of fact, the trial judge only considered Dr. Stoeckle’s opinion with respect to the degree of Galatis’s mental and cognitive impairment; she relied on the testimony of Dr. Whaley as to whether Galatis met “the necessary criteria for possessing testamentary capacity.”
The governing statute was repealed and replaced during the pendency of these proceedings (something that neither side addresses). However, both statutes require the proponent of a will to file a petition for probate. See G. L. c. 192, § 1, as in effect prior to St. 2008, c. 521, § 12 (requirements for filing of petition for probate); Marco v. Green, 415 Mass. 732, 738 (1993) (interpreting G. L. c. 192, § 1, as “requiring” the filing of a petition); G. L. c. 190B, § 3-402(a)(l), inserted by St. 2008, c. 521, § 9 (Massachusetts Uniform Probate Code) (requiring that petition for formal probate of will be filed requesting order “in relation to a particular instrument” [emphasis added]).
Our review of the Probate and Family Court docket reveals that on November 15, 2013, after judgment entered disallowing the February 9 will, a petition was filed to probate the February 1 document. The fate of that petition, which apparently was filed by Damaskos, is not before us.
We deny the contestants’ request for double costs, damages, and interest pursuant to Mass.R.A.P. 25, as appearing in 376 Mass. 949 (1979).