DALY, J.

The defendant conservation commission of the city of Norwalk appeals from the judgment of the trial court sustaining the appeal of the plaintiffs, Seaborne Herbert Tanner, Jr., et al. The plaintiffs had sought to obtain the defendant's approval for the construction of a single-family dwelling on their combined adjoining property. The commission claims that the trial court erred (1) in finding that it had acted without substantial evidence in its denial of the application, (2) in finding that all of the expert witnesses had agreed that the proposed residences would have no significant impact on area wetlands, and (3) in ruling that the agency is incapable of determining environmental impact without expert testimony. We find no error.

The following undisputed facts are relevant to this appeal. The plaintiffs are the owners of abutting land within a designated wetlands area located between Westview Lane and Oakhill Avenue in Norwalk. These lots were combined to create a 1.115 acre site and the plaintiffs applied for permission to construct a single-family residence thereon. The defendant is the duly designated municipal commission entrusted with enforcement of the Inland Wetlands and Watercourses Act, General Statutes 22a-36 through 22a-45, and the wetlands regulations of the town of Norwalk.

On May 29, 1984, a hearing was held on the application by the agency. The plaintiffs presented the testimony of two professional engineers, a soil scientist and a biologist. These were the only expert witnesses who offered evidence at the hearing. The commission claims that while Holt McCord, a professional engineer, and Bruce Lasky, a soil scientist, agreed that the single-family dwelling as proposed would have no significant impact upon the soil, wetlands or watercourse systems, the biologist, John Campbell, disagreed, hence providing substantial evidence for the board's refusal.

The trial court held that the experts stated that, in their opinions, the proposed house and driveway would have no significant impact on the wetlands. On the basis of this finding, the court sustained the appeal, determining that the commission had acted without substantial evidence to refute the plaintiffs' experts.

Our review of the record indicates that all the experts were in agreement that the proposed project could be built on the site. The commission created a disparity among the experts where none existed. Campbell stated that "the site has some wetland value, especially along the stream, but I do also believe that the stream should not be touched, and that the canopys of trees [around the stream fifty feet on either side] not be touched at all . . ." He subsequently restated his position, however, to the effect that the proposed house could be built without any adverse impact. It is clear from the trial court's determination that there was no disagreement among the experts that this was, in fact, the case.

The commission's first claim of error is that the trial court erred in finding that the commission had acted without substantial evidence in denying the plaintiffs' application for a single-family dwelling. We disagree.

At the outset, we note that the authority of an inland wetlands agency is extremely limited in that it can consider only matters that impact on designated wetlands areas. Connecticut Fund for the Environment, Inc. v. Stamford, 192 Conn. 247, 250, 470 A.2d 1214 (1984); Glendenning v. Conservation Commission, 12 Conn. App. 47, 52, 529 A.2d 727, cert. denied, 205 Conn. 802, 531 A.2d 936 (1987). The test that the reviewing court must apply in determining whether the commission's decision should be sustained is referred to as the substantial evidence rule. Huck v. Inland Wetlands Watercourses Agency, 203 Conn. 525, 552, 525 A.2d 940 (1987).

"`This so-called substantial evidence rule is similar to the "sufficiency of the evidence" standard applied in judicial review of jury verdicts, and evidence is sufficient to sustain an agency finding if it affords "a substantial basis of fact from which the fact in issue can be reasonably inferred. . . . [I]t must be enough to justify, if the trial were to a jury, a refusal to direct a verdict when the conclusion sought to be drawn from it is one of fact for the jury." . . . "The `substantial evidence rule' is a compromise between opposing theories of broad or de novo review and restricted review or complete abstention. It is broad enough and capable of sufficient flexibility in its application to enable the reviewing court to correct whatever ascertainable abuses may arise in administrative adjudication. On the other hand, it is review of such breadth as is entirely consistent with effective administration. . . . [It] imposes an important limitation on the power of the courts to overturn a decision of an administrative agency . . . and to provide a more restrictive standard of review than standards embodying review of `weight of the evidence' or `clearly erroneous' action. . . ."' (Citations omitted.) Lawrence v. Kozlowski, [ 171 Conn. 705, 713-14, 372 A.2d 110 (1976), cert. denied, 431 U.S. 969, 97 S.Ct. 2930, 53 L.Ed.2d 1066 (1977)]; Persico v. Maher, 191 Conn. 384, 409, 465 A.2d 308 (1983). The United States Supreme Court, in defining `substantial evidence' in the `directed verdict' formulation, has said that it `is something less than the weight of the evidence, and the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency's finding from being supported by substantial evidence.' Consolo v. Federal Maritime Commission, 383 U.S. 607, 620, 86 S.Ct. 1018, 16 L.Ed.2d 131 (1966); see Consolidated Edison Co. v. National Labor Relations Board, 305 U.S. 197, 229, 59 S.Ct. 206, 83 L.Ed. 126 (1938). `The reviewing court must take into account [that there is] contradictory evidence in the record . . . but "the possibility of drawing two inconsistent conclusions from the evidence does not prevent an administrative agency's finding from being supported by substantial evidence . . . ."' American Textile Manufacturers Institute, Inc. v. Donovan, 452 U.S. 490, 523, 101 S.Ct. 2478, 69 L.Ed.2d 185 (1981), quoting Consolo v. Federal Maritime Commission, [ 383 U.S. 607, 620, 86 S.Ct. 1018, 16 L.Ed.2d 131 (1966)]." Huck v. Inlands Wetlands Watercourses Agency, supra, 541-42.

In this case, the agency was cognizant of the holding in Feinson v. Conservation Commission, 180 Conn. 421, 429, 429 A.2d 910 (1980), that a lay commission acts without substantial evidence "when it relies on its own knowledge and experience concerning technically complex issues . . . ." The commission asserts that it was correct in disregarding the conclusion of two of the plaintiffs' experts while taking the recommendation of Campbell, that the stream should not be rerouted.

As stated previously, this reliance upon the testimony is misplaced, since an examination of the transcript and Campbell's report indicates that it was his opinion that construction of the proposed activity would have no adverse impact on the wetlands. The trial court did not hold that the evidence was insufficient on the basis of its improperly excluding proper evidence from its consideration. See Central Bank for Savings v. Planning Zoning Commission, 13 Conn. App. 448, 458, 537 A.2d 510 (1988). Rather, this is a case where the trial court recognized that the commission, in dealing with complex issues, ignored the testimony of the expert witnesses and relied solely on their own insight. Feinson v. Conservation Commission, supra, 428. While we recognize that an administrative agency is not required to believe any of the witnesses, including expert witnesses; Manor Development Corporation v. Conservation Commission, 180 Conn. 692, 697, 433 A.2d 999 (1980); it must not disregard the only expert evidence available on the issue when the commission members lack their own expertise or knowledge. We find there was no substantial evidence since there was an absolute disregard of the unanimous contrary expert opinion.

In finding that there was no substantial evidence, we have considered the remaining two claims of error and find them to be without merit. Accordingly, we hold that the trial court was correct in its determination that the action of the defendant should be set aside and that the plaintiffs' appeal sustained.