(dissenting) — It is my opinion that the learned trial judge properly exercised his discretion in dismissing the charges against defendant. Accordingly, I dissent.
I do agree with the following holdings of the majority:
(1) Actual possession of the marijuana by defendant *305was not established as a matter of law. The case of State v. Callahan, 77 Wn.2d 27, 459 P.2d 400 (1969) compels such a conclusion.
(2) It was within the broad discretion of the trial court to exclude the many packages of alleged marijuana, the contents of which had not been tested, for the reasons stated by the majority.
However, I depart from the conclusion of my colleagues that the fingerprint evidence was proper, since the state failed to prove specifically that the contents of the packages which bore the defendant’s fingerprints contained marijuana.
A broad discretion rests in the trial court with reference to the admission of this type of evidence, and particularly should this be so where there is a defect in either the identification of the evidence or a defect in the chain of evidence. I would hesitate to find an abuse of discretion in the face of an almost unbelievably careless job done by the police in failing to keep track of the tested packages, or of the location where the tested packages were found on the premises, and in failing to test the packages containing the fingerprints or to identify the locus of those packages.
The broad discretion which the trial court has in this area of criminal jurisprudence was discussed in State v. Brooks, 3 Wn. App. 769, 479 P.2d 544 (1970) and State v. Golladay, 78 Wn.2d 121, 470 P.2d 191 (1970). Under the principles of those cases, I do not find an abuse of discretion in this case.
However, even if it were assumed that the fingerprint evidence should have been allowed, I still would agree with the trial court that the state failed to make a prima facie case that defendant had dominion and control over the premises in which the marijuana was found. In my opinion, the evidence that defendant had handled one or more of these packages is no more probative of constructive possession than it would be of actual possession. State v. Callahan, supra. The momentary handling principle discussed in Callahan should apply with equal force to a case grounded *306in constructive possession as it does in actual possession. For this reason the error, if any, in excluding the fingerprint evidence was harmless.,
Where I must substantially depart from the majority opinion is its holding that the state made a prima facie case that defendant had dominion and control of the premises where the marijuana was found.
My disagreement with the majority stems from its assumption that circumstantial evidence pointing to the fact that defendant may have previously occupied the premises, coupled with circumstantial evidence of his present occupancy, represents a sufficient prima facie showing that he had dominion and control of the home.
This assumption is faulty, in my opinion. In State v. Callahan, supra, the defendant (1) admitted actual occupancy of the houseboat at the time of his arrest and for 2 or 3 days prior to that time, (2) admitted that he knew the drugs were there, (3) admitted that his personal property was there, (4) admitted that he had handled the drugs, and (5) was within a few feet of the drugs at the time they were discovered.
If the evidence of actual occupancy shown in Callahan is insufficient in law to establish dominion and control, then the circumstantial evidence pointing to defendant’s occupancy of the premises in this case is clearly insufficient to prima facie establish his dominion and control of the premises.
In the case at bar, there is not one shred of evidence that defendant had any “legal” connection with the premises, other than as a past occupant, or at the time of the search as a musician who rehearsed at the premises with his group. Furthermore, the current utility bill for the premises covering the month in which the search and arrest occurred was posted to one “Doug Taylor.”
As in Callahan, there was no testimony that defendant was a tenant, nor that he participated in paying the rent or maintained the place as his residence. There was no testimony that he ever remained at the premises overnight.
*307Furthermore, all of the cases considered by the Supreme Court and discussed in Callahan, where constructive possession was allowed, have involved proof of some “legal” connection to the premises, such as tenancy, or have involved evidence of residency tantamount to tenancy.2 Such has also been the thrust of prior decisions of this court. State v. Tretton, 1 Wn. App. 607, 464 P.2d 438 (1969); State v. Werry, 6 Wn. App. 540, 494 P.2d 1002 (1972). Such an application of the constructive possession rule is, I think, entirely proper and necessary to insure against miscarriage of justice.
There are today many instances of young people entering into loose living arrangements, particularly around colleges and universities. Short-time occupancies occur with sufficient frequency that I am loath to permit strained inferences of dominion and control from such types of occupancies.
To impute, or allow juries to find, dominion and control so as to establish constructive possession on the paucity of evidence presented in this case would, in my opinion, unduly threaten innocent persons with drug charges properly attributable to others. I believe this to be the real thrust of Callahan, a thrust with which I agree.
I would restrict application of the constructive possession theory to those instances in which (1) a tenancy is established, (2) a permanent type of residency indicative of a tenancy is established, or (3) where dominion and control may reasonably be inferred from circumstances other than occupancy alone. Since none of these factors was established prima facie, I think the trial court was correct in dismissing the information.
I have another reason for my view. The constructive possession theory of criminality is not strictly a factual theory. Instead, it is partly a legal fiction which is utilized *308to impose criminality on those who are shown to have a sufficiently close “legal” connection to the locus where drugs are found, which, for policy reasons, should be considered the legal equivalent of actual possession. Its application should, therefore, be cautiously and strictly considered, so that miscarriages of justice are not likely to occur. Under these circumstances, I think the trial court should have some discretion, which strict application of the circumstantial evidence rule enunciated in State v. Randecker, 79 Wn.2d 512, 487 P.2d 1295 (1971) does not allow, to determine whether the “legal” connection of defendant to the locus is sufficiently shown so as to prevent a possible miscarriage of justice.
I am in complete agreement with the statement of the Supreme Court in State v. Golladay, supra, citing with approval at page 130 from State v. Weaver, 60 Wn.2d 87, 88, 371 P.2d 1006 (1962): “While a conviction may be sustained solely on circumstantial evidence, the circumstances proved must be unequivocal and inconsistent with innocence.” (Footnote omitted.)
For these reasons, and because there was not one shred of evidence, circumstantial or otherwise, which bore directly or indirectly on defendant’s control of the premises as distinguished from his occupancy, I would affirm the judgment of dismissal.
“Possession of the key to the premises was held sufficient in State v. Mantell, 71 Wn.2d 768, 430 P.2d 980 (1967), but there was also direct evidence that defendant was participating in preparation of drugs for sale.