McGOWAN, Circuit Judge:

R. J. Mazza died a resident of Maryland, leaving a will which named appellant — his wife, Ennis Mazza — executrix and sole beneficiary, and which made no provision for the payment of the federal estate tax. The decedent and appellee — his sister, Olga Mazza — held real property in the District of Columbia as joint tenants with rights of survivorship. That property passed to appellee by operation of District of Columbia law, and was included in R. J. Mazza's taxable estate under § 2040 of the Internal Revenue Code of 1954. Appellant paid the entire federal estate tax, and sought to compel contribution from appellee for that portion of the tax attributable to the real estate in the District of Columbia. The District Court granted appellee's motion for summary judgment on the grounds that the question of apportionment was governed by the local law of the District of Columbia, and that under that law estate taxes are to be paid out of the residuary estate. Appellant argues that (1) the court should have looked to the law of Maryland for the rule governing apportionment and (2) even if the law of the District of Columbia applies, the court misconstrued that law. We reverse on the basis of the choice of law issue and do not reach the question of interpretation.

Congress expressly provided for apportionment of the federal estate tax in certain situations not relevant to this case. In all other instances the decedent's personal representative is responsible for actual payment, but state law determines the impact of the tax upon those receiving property includible in the taxable estate. Riggs v. Del Drago, 317 U.S. 95, 63 S.Ct. 109, 87 L.Ed. 106 (1942). Maryland has a statute requiring pro rata apportionment of the tax unless the testator makes specific provision for its payment. The law in the District of Columbia is somewhat less clear, but the District Court concluded that apportionment is not required and that taxes are to be paid from the residuary estate. Inasmuch as it is unnecessary to reach appellant's challenge to this interpretation, we assume for the purposes of this discussion that the District Court accurately ascertained the local law.

I

The conflict of laws problem arises because both Maryland and the District of Columbia have significant contracts with this controversy and their rules of law differ. The separate interests and policies of the two jurisdictions have been drawn together by the broad cast of the federal estate tax net, which assesses a single lump sum liability for all assets within the taxable estate regardless of location. Our task is to sort out these diverse elements in an attempt to determine the relationship of each jurisdiction to the controversy, and to evaluate the interest of each in the application of its own rule of law.

The foregoing statement of the problem indicates that we adopt in this case the "interest analysis" approach initially employed by this court in the area of tort law in Tramontana v. S. A. Empressa De Viacao Aerea Rio Grandense, 121 U.S.App.D.C. 338, 350 F.2d 468, 471-473 (1965), and extended to the area of contracts by cases such as Fox-Greenwald Sheet Metal Co. v. Markowitz Bros. Inc., 147 U.S.App.D.C. 14, 452 F.2d 1346, 1353-1354 (1971). Although the subject of this controversy is somewhat different, the reasoning of those cases seems equally applicable here.

Although the general approach to be employed is thus established, its application to the issue of whether the situs should adopt the domicile's apportionment statute is a question of first impression in this jurisdiction. Only a few other courts have considered the problem, and those that have are evenly divided between the law of the situs and that of the domicile. The two cases which best illustrate the opposing views are Isaacson v. Boston Safe Deposit Trust Co., 325 Mass. 469, 91 N.E.2d 334 (1950), and Doetsch v. Doetsch, 312 F.2d 323 (7th Cir. 1963). For reasons developed hereinafter, we find the reasoning of the Seventh Circuit in Doetsch persuasive in principle, and also consistent with the choice of law principles of this jurisdiction.

II

In Doetsch the decedent died domiciled in Arizona, leaving his widow as income beneficiary of an inter vivos trust of property located in Illinois. The corpus of the trust was included in the taxable estate, and since the will was silent on this point, the tax was paid from the income of the trust; subsequently, the widow sought to charge the remaindermen with a pro rata share of the tax. Neither state had an apportionment statute, and there were no Illinois decisions on the conflicts question. The jurisdiction of the federal courts was based on diversity of citizenship, and Chief Judge Hastings, writing for the court, decided that an Illinois court would apply the local law of Arizona. He then observed that Arizona had not decided the question and concluded that Arizona would adopt apportionment as the better rule of law.

The court made three points in its discussion of the choice of law question. First, reference by all jurisdictions in which the decedent left property to the law of the decedent's domicile insures uniform treatment of all those receiving property from the decedent's taxable estate. Second, this question is similar to other problems relating to the administration of estates and to determination of intent which are governed by the law of the domicile. Finally, the decedent's domicile is usually the jurisdiction concerned with the protection of the decedent's widow and children, and deference to that state's policy in such matters is appropriate.

In determining Arizona's interest in protection of the residuary beneficiaries, the Doetsch court was without the benefit of any pronouncement of Arizona policy. Our case is much easier because enactment of the apportionment statute is a clear expression of Maryland's approach to the question. Although the statute does not explain its purpose, it seems intended to protect residuary beneficiaries from the untoward effects of unforeseen taxes. This decision may well be premised on the conclusion that residuary beneficiaries are likely to be intended principal beneficiaries, and that a failure to provide for payment of taxes will almost certainly be an oversight.

Since the residence of the decedent will commonly be the residence of the spouse and dependent children, and since such family members are usually the principal beneficiaries, Maryland, as the decedent's residence, has a dominant interest in protection of the principal beneficiary. Further, the approach adopted by Maryland seems clearly suited to serve that purpose. As the federal estate tax has become more substantial, methods of disposing of property outside of the testamentary estate have become more sophisticated, and the definition of taxable estate has consequently been expanded to include many types of property not subject to administration by the decedent's personal representative. These changes increase the probability that property will be included in the taxable estate unexpectedly, and that the testamentary scheme will be thereby distorted. Unlike pro rata apportionment, which can at worst reduce by a percentage some assets which the testator expected would be transferred in their entirety, payment of the taxes from the residuary estate can consume that estate entirely, perhaps leaving an intended principal beneficiary with nothing at all.

In addition to this local concern with the residuary beneficiaries, the Doetsch opinion identified an interest in uniformity of treatment of beneficiaries within a single estate. If a decedent leaves property in several states, and if each situs applies its own law, some of the recipients may be required to contribute to payment of the federal estate taxes while others are not. Different treatment for persons similarly situated with respect to the estate is an anomalous result which can be avoided if all jurisdictions refer to the law of the domicile.

III

Against these interests in the application of Maryland's apportionment statute, we must balance the interests of the District of Columbia in the application of its own law. The District Court seems to have based its opinion primarily on considerations of extraterritoriality. The court issued no opinion in this case, but its fifth Conclusion of Law reads as follows:

5. Under the District of Columbia law, an apportionment statute in the place of the decedent's domicile has no extra territorial [ sic] effect against the surviving joint tenant of real estate located in the District of Columbia.

In following this approach the District Court seems to adopt the rationale of Isaacson v. Boston Safe Deposit Trust Co., 325 Mass. 469, 91 N.E.2d 334 (1950), the leading case for application of the law of the situs in cases of this sort.

Isaacson was a suit by the executor of a will against the trustee of a trust to charge the trustee with a pro rata share of the federal estate tax. The decedent died domiciled in Maine, and the situs of the trust was in Massachusetts. Maine had an apportionment statute, but Massachusetts, the forum, did not. The court noted that the trust was created under Massachusetts law while the decedent was domiciled in Massachusetts, and the trust property was in the state. The trustee was a Massachusetts corporation, and the persons who succeeded to the decedent's interest in the trust did so by the terms of the trust and not under the will probated in Maine. For these reasons, the court concluded "that to apply the Maine statute in this case would be to give to that statute extraterritorial effect contrary to first principles." The court then cited without explanation a long series of cases, most of which deal with attempts by forum states to apply the local law of the forum to controversies in which the forum had no interest.

The reasoning of Isaacson and the District Court is correct when applied to the question of whether an apportionment statute in the domiciliary state must be applied by a forum which has jurisdiction as the situs of real property. The Maryland statute does not, however, apply to this controversy by force of Maryland law. As situs of the property, the District of Columbia could apply its local law without running afoul of the constitutional strictures of Due Process or Full Faith and Credit.

This conclusion is, however, only the first step and does not reach the choice of law question at all. To the extent that the District Court felt compelled by territorial considerations to apply the law of the District of Columbia, the court failed to distinguish between an attempt by a domiciliary state to apply its local law regarding tax apportionment to land located in another jurisdiction, on the one hand, and the decision of a situs forum to defer to the local law of the domicile, on the other. The former would be an attempt to extend the effect of the forum's own local law to property not present in the jurisdiction and would arguably give unjustifiable extraterritorial effect to the law of the forum.

The decision of a situs to apply the law of the domicile is, however, simply a determination that the forum's own choice of law policies are best served by reference to foreign law. Whenever a forum under no compulsion to do so elects as a matter of conflicts policy to apply the law of another jurisdiction, that law is given "extraterritorial effect" in the sense that such effect is present here. But since there are no incursions upon sovereignty in the application of foreign law, this "effect" is in no sense a barrier to such deference. The relevant inquiry focuses on the relationships of the two jurisdictions to the controversy, the interests involved, and whether application of foreign law would offend a strong and clearly defined local policy. Objections to application of foreign law would be justified if this analysis were to disclose that important interests of the forum would be sacrificed to advance equal or lesser interests of another jurisdiction, but resolution of that question is not advanced by reference to extraterritoriality.

IV

Although we do not attempt to resolve our choice of law problem simply by casting it within the mold of one or another of the traditional conflicts categories, it is still appropriate to determine whether the principles and policies which forged those rules are applicable to our situation. Concerns with the stability of use of, and marketability of title to, land were bases of the traditional conflicts rule that the law of the situs governs questions of succession to land, and allocation of the federal estate taxes is at least a related issue.

In respect of such considerations, however, it is possible to distinguish between questions of succession to land and those concerning apportionment of estate taxes against the land. The question of apportionment could affect title to the land only indirectly at best, and would in any event affect only a portion of the value. Unlike questions relating to the validity of title, the issue of tax liability is one of short duration. The attenuation of this relationship suggests that a principal reason for the concern with stability of title — the danger of third-party reliance on the law of the situs — is insubstantial with regard to the responsibility for estate taxes. Since it seems that the interests underlying the traditional rule are not involved in this case, that rule casts no weight into the balance.

We must further determine whether the assumed District of Columbia rule of nonapportionment reflects some strong policy or protects a significant interest. Initially it should be noted that there has been no clear explication of the law of the District of Columbia on this point. If the law is nonapportionment, it is by virtue of the common law rule that taxes are expenses to be paid out of the residuary estate. That rule developed on the basis of the presumed intent of the testator. However, as noted previously, the entire area of estate planning and estate taxation has undergone substantial change in recent years. In the midst of that change the District of Columbia has at most retained the old rule with no apparent reexamination of the rule's relationship to the policies which originally justified its adoption. Nor has any new justification for the rule been suggested in other cases or by the parties in this proceeding. Although we do not decide that there is no longer any basis for the rule of nonapportionment in the District of Columbia, neither do we discern a strong local policy requiring application of the rule to the facts before us.

V

On balance we conclude that Maryland is the jurisdiction with the most significant interest in this controversy, and that application of the Maryland statute would best serve the various policies to be considered. The statute is a clear expression of a state policy to protect persons such as appellant. The law, and consequently its basis, are much less clear in the District of Columbia. Further, reference to the law of the domicile will result in uniform treatment of all beneficiaries of an estate. Absent an identifiable interest in the application of the substantive law of the District of Columbia, it is unimportant that this decision gives the Maryland statute effect where not constitutionally compelled. Within our federal system deference to the law of the jurisdiction of dominant interest should not be viewed as a threat to the sovereignty of the forum. Such action should promote a spirit of cooperation and should help insure that other jurisdictions will be sensitive to the forum's policies.

Accordingly, we reverse and remand for further proceedings consistent with this opinion.

It is so ordered.