The opinion
L. HAND, Circuit Judge
(dissenting).
There can be no doubt that somebody made a profit when the bonds were bought in at a discount; and I think that that'profit was made “from sources within the United States.” § 119(a) of the Revenue Act of 1934, 26 U.S.C.A. Int.Rev.Acts, page 709. If a non-resident obligor makes a bond payable in this country, he must bring his funds into the country to pay it; if he buys it in at a discount, that may not relieve any of his assets then within the United States, but it does relieve him of the duty of bringing assets into the United States when the bond falls due. There seems to me to be no substantial difference between that and relieving assets already within the United States. It does not of course follow that the taxpayer at bar should be taxed as though this profit were its own, and I feel the force of my brothers’ reasoning to the contrary. I assume—because the taxpayer has the burden of showing the contrary—that the only persons who had any interest in its assets were the four “adherents”—“Cosach in Liquidation,” Antofagasta, Lautaro and Anglo-Chilena. The taxpayer’s obligation, though it was in form to pay the principal and interest of the bonds, was so limited that the burden of it could not be appraised, and there is no basis for estimating its profit unless we may look to the assets and undertakings of the “adherents” and treat them and the taxpayer as one.
The relations between them were that the taxpayer collected the earnings, paid the royalty to the Chilean government and distributed the surplus to the four “adherents” in proportions which varied from year to year. It had promised to pay the bonds only to the extent of the earnings; but the “adherents” promised without limitation to pay them so far as the taxpayer did not; and that promise was joint and several. The situation therefore was that any payments made by the taxpayer relieved the “adherents’ ” general assets quite as much as though they paid themselves, and that the payments came out of earnings which would otherwise be distributed to them. But we can tell neither how the earnings would be distributed, nor how the eventual burden of the bonds would have been distributed, even though *145 that would be in the same proportion as the earnings were distributed. For this reason I do not see how any particular “adherent” could be taxed upon its aliquot part of the profit; and this is equally true if the proper distribution of the burden of the bonds among the “adherents” be in some other proportion than their share of the earnings. Thus, although the “adherents” are the proper persons to tax, it is impossible to tax them severally. It is, however, possible to tax them collectively, leaving the distribution of the burden to them. My brothers suggest that perhaps the sum of their several taxes is less than the tax against the taxpayer, but to this it seems to me a sufficient answer that they have not proved that that is so, and so far as I can see they could not do so. The only possible objection therefore is that the tax was laid upon the taxpayer and not against the “adherents” collectively. That seems to me to be merely matter of form; they were liable collectively; they held out the taxpayer to act for them collectively, and the consequences of taxing it in the end are the same as taxing them severally, which is impossible. I do not think that either the taxpayer or they have any just ground of complaint that the tax is collected in this, the only feasible way.