(dissenting). We dissent. We agree that defendant was not entitled to a suppression hearing and has failed to preserve his challenge to the presence of the police and dog in the common hallway (see, CPL 710.70 [3]; cf., People v Lott, 102 AD2d 506). We disagree that the dog sniff did not constitute an illegal search.
At the outset, it is important to stress that the dog sniff was conducted at the door of defendant’s apartment at the specific request of a police officer who was searching for evidence of a crime. The only case directly addressing such a procedure conducted at a doorway is the Second Circuit Court of Appeals opinion in United States v Thomas (757 F2d 1359, cert denied sub nom. Wheelings v United States, 474 US 819) which held that a dog sniff at a dwelling is a warrantless search in violation of the Fourth Amendment. Unlike the majority, we find the Thomas rationale most persuasive. Here, as in Thomas, and unlike the cases upon which the majority relies, the dog intruded upon defendant’s residence in which defendant had not only a legitimate, but also a heightened expectation of privacy (see, Payton v New York, 445 US 573, 590). This is because "physical entry of the home is the chief evil against which the wording of the Fourth Amendment is directed” (United States v United States Dist. Ct., 407 US 297, 313; see also, People v Levan, 62 NY2d 139, 144).
The majority holds these principles do not apply because "the dog did not enter defendant’s residence but merely sniffed the air outside the apartment.” The court in Thomas believed otherwise. A trained dog, like other devices used by the police (e.g., magnetometers or "spike-mikes”), may detect the contents of a private enclosed space, here defendant’s residence, that the police could not otherwise obtain from the use of their own sense of-smell. "In practical effect there is no difference between the emanations of odor sniffed by the dog and the sound vibrations sensed by such devices. Both originate from inside a private area and travel beyond its perimeters” (People v Price, 54 NY2d 557, 565 [Meyer, J., concurring]). Moreover, at the time of the dog sniff the apartment was vacant and there was no indication that any odor of illegal drugs was detectable to human olfactory senses. The precise reason the dog was brought to the apartment was to see if its sensitive nose could detect the odor of narcotics.
The court in Thomas was acutely aware of the sanctity of one’s dwelling (supra, at 1366-1367): "Thus, a practice that is not intrusive in a public airport may be intrusive when *94employed at a person’s home. Although using a dog sniff for narcotics may be discriminating and unoffensive relative to other detection methods, and will disclose only the presence or absence of narcotics, see United States v Place, 103 S. Ct. at 2644, it remains a way of detecting the contents of a private, enclosed space. With a trained dog police may obtain information about what is inside a dwelling that they could not derive from the use of their own senses. Consequently, the officers’ use of a dog is not a mere improvement of their sense of smell, as ordinary eyeglasses improve vision, but is a significant enhancement accomplished by a different, and far superior, sensory instrument. Here the defendant had a legitimate expectation that the contents of his closed apartment would remain private, that they could not be 'sensed’ from outside his door. Use of the trained dog impermissibly intruded on that legitimate expectation.” Although the Thomas decision was based on interpretation of the Fourth Amendment of the US Constitution, its holding is also supported by interpretation of article I, § 12 of the NY Constitution which affords greater protection under more exacting standards (see, People v P. J. Video, 68 NY2d 296, cert denied 479 US 1091).
We all agree that absent information obtained from the dog sniff there was no probable cause to support the search warrant (see, People v Tambe, 71 NY2d 492, 505; People v Edwards, 69 NY2d 814, 816). This is so not only because there is no assurance of the reliability of the hearsay information from Gellart and Osgood upon which Gramaglia relied in seeking the warrant (see, People v Hendricks, 25 NY2d 129, 134-137; People v Simon, 107 AD2d 196, 202), but also because this information was stale since Gramaglia failed to present the Magistrate with any information concerning a particular drug transaction in which defendant was involved near the time the warrant was issued (see, Sgro v United States, 287 US 206; People v Christopher, 101 AD2d 504, 527, revd on other grounds 65 NY2d 417, rearg denied 65 NY2d 1054).
Lastly, we note our disagreement with the concurring opinion which argues that although a dog sniff is a search it may be justified on less than probable cause. There are rare instances when this position may be appropriate but this is not one of them. Here, the serious invasion of defendant’s privacy by the dog sniff at the front door of his residence is in no way analogous to a stop and frisk (see, Terry v Ohio, 392 US 1), or an administrative search (see, Camara v Municipal Ct., 387 *95US 523) or a border search (see, United States v Brignoni-Ponce, 422 US 873) which may be justified on less than probable cause. There are no concerns implicated here involving an officer's safety or administrative expedience which are at the heart of these narrow exceptions to the general probable cause requirement under the Fourth Amendment. It is clear that Terry v Ohio (supra) and the cases that followed it permit only brief investigative stops and extremely limited frisks based on reasonable suspicion. "They do not provide the police with a commission to employ whatever investigative techniques they deem appropriate” (United States v Place, 462 US 696, 714 [Brennan, J., concurring]).
As noted at the outset, the police use of the trained dog in this case was not accidental. Rather, the dog sniff here was conducted at the specific request of a police officer who was searching for evidence of a crime behind the closed door of defendant’s residence. Under such circumstances we agree with the Thomas rationale that this constitutes a search which must be supported by probable cause before, not after, the dog conducts the sniff. The concurrence permits a dog sniff to be conducted on less than probable cause and then uses the information from the sniff, to find probable cause to support the search warrant for defendant’s residence. This reasoning is circular. The critical threshold issue is the intrusiveness of the search (see, United States v Place, supra, at 722 [Black-mun, J., concurring]). If a dog sniff of a defendant’s residence constitutes a search, as the concurrence agrees it does, then it must be supported by probable cause prior to the sniff. The sniff, as in the case of any search, cannot be justified by what it turns up. It is fundamental law that a search is good or bad at its inception and cannot be justified by what is thereafter discovered (United States v Di Re, 332 US 581; Johnson v United States, 333 US 10).
Accordingly, the warrants issued for defendant’s apartment were invalid. Therefore, we would reverse defendant’s conviction and dismiss the indictment.
Boomer, J., concurs with Denman, J. P.; Lawton, J., concurs in a separate opinion; Green and Davis, JJ., dissent in an opinion by Green, J.
Judgment affirmed.