(dissenting). The majority properly recognize the “well-established principle” that a property owner’s self-imposed hardship cannot serve as the basis for obtaining a variance, as well as the fact that Brolly knowingly created the nonconformity when he subdivided his property and transferred one of the newly created lots, leaving his remaining lot with insufficient frontage. They attempt to save this situation from application of that principle by emphasizing, as did the trial judge, Brolly’s good faith — he “did not create the [nonconforming] lot in anticipation of being granted a variance,” and he took all the necessary steps to bring the lot into conformance with the frontage requirement, but was thwarted by an event beyond his control, the MDC taking. I am unpersuaded that either the law or the facts of this case support their effort.
The instant case is essentially indistinguishable from Gordon v. Zoning Bd. of Appeals of Lee, 22 Mass. App. Ct. 343 (1986). There, we reversed the grant of a variance for a “pork chop” shaped lot with inadequate frontage which had been created as a remnant when the owner sold a portion of his property for its subdivision into several conforming lots. We did so on the basis of a long line of decisions holding that “the creation by conveyance of an unusually shaped, nonconforming parcel under then existing zoning regulation does not entitle one to a variance.” Id. at 351. That oft-repeated holding applies equally to Brolly’s situation.
The Gordon opinion also took note of the fact that, at the time the owner had sold his land and created the nonconforming parcel, he had no further plans to use it except as a wood lot; and that he attempted to convey it only six years later, when an “unexpected opportunity” to sell arose. Id. at 345. An additional lesson of Gordon, therefore, is that the absence of an intention on the part of the owner, at the time he knowingly creates an unusually shaped, nonconforming parcel, to obtain a variance for the parcel is irrelevant, as the express language of G. L. c. 40A, § 10, would suggest.
The facts here do not appear to justify the majority’s conclusions that Brolly did everything he could to cure the unfavorable characteristics of the locus which he had created and that his ultimate inability to proceed was of the MDC’s, not his, making. Most obviously, he could have delayed any sale (and nothing in the record suggests any reasons why he could not *9have) until he had procured the McNeil property.1 He could also have awaited the actual clearance of title on McNeil’s lot before creating the nonconformity by his August 7, 1992, sale of a portion of his lot.2 In this connection, I do not agree that the underlying facts justify the trial judge’s and the majority’s finding that the failure to implement the land swap transaction and the “hardship” that rendered Brolly’s lot unbuildable were attributable to the ultimate MDC taking, which occurred over a year and one-half after the transaction was agreed upon.3 The parties’ expectations seem rather to have foundered because of the perceived title or encumbrance problems and their failure or inability to resolve them, despite many months of opportunity *10to do so. (In any event, the finding is irrelevant to the underlying issue; see note 1, supra.)
Finally, I view the authorities on which the majority principally rely — Paulding v. Bruins, 18 Mass. App. Ct. 707 (1984), and Smith v. Zoning Bd. of Appeals of Norwalk, 174 Conn. 323 (1978) — as inapposite. In neither case had the owner himself (unlike Brolly) created the circumstances that rendered the locus nonconforming.4 I find more authoritative, in addition to Gordon, the decision in Tsagronis v. Board of Appeals of Wareham, 415 Mass. 329 (1993). There, the Supreme Judicial Court held that the “shape of the land” criterion of G. L. c. 40A, § 10 (which would have to be satisfied to warrant a variance here, because there is nothing in the record suggesting uniqueness of soil conditions or topography), is not met merely because the shape of the particular lot as subdivided is such as to have insufficient frontage. “[Fjailure to meet dimensional requirements does not satisfy the odd shape criterion of the statute.” Id. at 332 (citations omitted). Accord Warren v. Zoning Bd. of Appeals of Amherst, 383 Mass. 1, 11 (1981); Arrigo v. Planning Bd. of Franklin, 12 Mass. App. Ct. 802, 804 (1981); Guiragossian v. Board of Appeals of Water-town, 21 Mass. App. Ct. 111, 116 (1985); Gordon, 22 Mass. App. Ct. at 351; DiCicco v. Berwick, 27 Mass. App. Ct. 312, 313-314 (1989); Karet v. Zoning Bd. of Appeals of Worcester, 27 Mass. App. Ct. 439, 440 (1989); Feldman v. Board of Appeal of Boston, 29 Mass. App. Ct. 296, 297 (1990).
Accordingly, I agree with the appellant Adams and would reverse the judgment below and order entry of a new judgment annulling the decision of the zoning board of appeals as in excess of its authority. See Warren v. Zoning Bd. of Appeals of Amherst, supra at 12-13.
The majority’s observation that, even if Brolly had waited to sell until he had acquired the McNeil property, the taking would have occurred in any event and was the real cause of Brolly’s “hardship” misses the point. It was Brolly’s obligation to demonstrate satisfaction of all the G. L. c. 40A, § 10, prerequisites in order to obtain a variance. While relevant to the “substantial hardship” requirement, the fact of the taking was irrelevant to the separate condition of unique “soil conditions, shape, or topography.” The statutory prerequisites for a variance under G. L. c. 40A, § 10, are conjunctive and not disjunctive, so that failure to satisfy even one defeats the variance. See Gordon, 22 Mass. App. Ct. at 348, 349 n.6.
The record is silent regarding the precise nature of McNeil’s title problems or the reasons for his inability to clear the title of “mortgage interests.” Nor do we know why Brolly could not have himself “cleared” the title or taken subject to the “mortgage interests,” with some form of title insurance or indemnity agreement. (We note, as to these questions, that almost nine months passed between the time of the Brolly-McNeil land swap agreement and the March, 1993, receipt of the MDC’s notice of “a possibility of a taking by eminent domain,” with no record explanation of what, if anything, Brolly and McNeil were doing to resolve the title problems during that lengthy period.) The record is also silent as to the reasons Brolly proceeded to sell part of his then-conforming lot virtually on the heels of obtaining constructive approval of his subdivision plan, and presumably with knowledge of the existing title problems as well as the zoning consequences of his sale. On its face, his situation appears indistinguishable from that in Gordon, supra at 350, where the owner had created the nonconforming condition “for [his] financial gain.” In the light of these several critical factual uncertainties, it seems inappropriate to describe Brolly’s as an “exceptional case.”
I note that G. L. c. 40A, § 10, does not identify an eminent domain taking as one of the circumstances that can give rise to the requisite hardship justifying variance relief. Even if it were, it would not be one that “relat[es] to the soil conditions, shape, or topography” uniquely affecting the particular property (see note 1, supra). It is therefore questionable whether the MDC taking was a relevant analytical factor in evaluating Brolly’s entitlement to a variance in the instant circumstances, which involved only dimensional nonconformity.
Paulding was the basis for the favorable decisions on the variance by the zoning board and the trial judge. The Paulding case is further distinguishable because it involved a lot created before the adoption of zoning. As pointed out in DiCicco v. Berwick, 27 Mass. App. Ct. 312, 314-315 (1989), no appellate decision has approved the grant of a variance to a lot that violated zoning frontage requirements at the time of its creation, and a long line of cases has disapproved of such variances.