delivered the opinion of the court:
The suit is to recover taxes alleged to have been illegally assessed and collected, the particular tax involved being the 4 per cent additional or war income tax imposed by section 4 of the revenue act of October 3, 1917, .40 Stat. 300, as affected in its assessment by section 1211, which amended the act of 1916 by adding thereto section 29.
*119The facts as found by the court are in accordance with a stipulation by the parties and the point at issue, which is as to the proper method of computing the tax, is quite clearly made to appear by Findings XI and XII, which set out an exemplification both of the method adopted and followed by the Revenue Bureau and that contended for by the plaintiff. The difficulty arises by reason of the fact that the plaintiff company is not a calendar year corporation but a fiscal year corporation, whose fiscal year began with June 1 and ended with May 31. The issue is, therefore, as to the 4 per cent tax assessed and collected for the five months of plaintiff’s fiscal year 1917 which fell within the calendar year 1917.
It seems scarcely necessary to quote in full all the provisions of the revenue acts bearing upon the question or to follow in detail the elaborate arguments on both sides. The 2 per cent income tax was in force in 1916 as well as 1917. The 4 per cent additional income tax provided by section 4 of the act of 1917 was a “ like tax ” of 4 per cent, in addition to the 2 per cent, but was specifically upon “the income received in the calendar year nineteen hundred and seventeen ” made more definite in its application by the provision following, that “ if it has fixed its own fiscal year, the tax imposed by this section for the fiscal year ending during the calendar year nineteen hundred and seventeen shall be levied, assessed, collected, and paid only on that portion of its income for such fiscal year which the period between January first, nineteen hundred and seventeen, and the end of such fiscal year bears to the whole of such fiscal year.” Section 1211 added to the act of 1916 by way of amendment six sections, including section 29, which provided that “ in assessing income tax the net income embraced in the return shall also be credited with the amount of any excess-profits tax imposed by act of Congress and assessed for the same calendar or fiscal year upon the taxpayer.”
There is no question as to the amount of excess-profits tax to be credited, there is no question as to the amount of net income for plaintiff’s fiscal year ending May 31, 1917, and parties are in agreement as to the computation of the 2 per cent tax which was of course upon the whole net income for *120the fiscal year after crediting the excess-profits tax. But in computing the 4 per cent tax for the 5 months of plaintiff’s fiscal year 1917 falling within the calendar year 1917, different methods of computation aré resorted to in this, that the Bureau of Internal Bevenue first takes plaintiff’s net income for its entire fiscal year, deducts therefrom the excess-profits tax and computes the 4 per cent tax on five-twelfths of the remainder, resulting in a tax of $64,546.41; while the plaintiff, as the basic sum of its computation, takes five-twelfths of the plaintiff’s income for the fiscal year, deducts therefrom the excess-profits tax and computes the 4 per cent tax upon this remainder, resulting in a tax of $48,458.76.
It seems to us that for at least one very apparent reason the computation of the Bureau of Internal Bevenue is upon an erroneous basis. When it took the entire income of the plaintiff for its fiscal year 1917 and deducted therefrom the excess-profits tax and computed the 4 per cent tax on five-twelfths of the remainder, it carried into its computation in its basic figures seven-twelfths of plaintiff’s net income for its fiscal year in question which had fallen within the calendar year 1916, and as to which year the 4 per cent tax was not applicable. This apparent result from this method of computation is disputed, but it seems to us quite clear that the ultimate effect of the method adopted is to tax a portion of plaintiff’s net income which was not in fact taxable.
Defendant’s counsel, if we understand correctly, contends that there is no provision for assessing the 2 per cent tax by one method and the 4 per cent tax by another method, and that the language of section 29, in its reference to “ the net income embraced in the return ” requires the use of the same basic figures in both computations. The theory apparently is that since the taxpayer returned its net income for the full fiscal year upon which, after the proper credit, the 2 per cent tax was assessed, the language quoted requires the use of the same “ income ” as the basis of computation in determining the 4 per cent tax. We see no force in the argument or basis for the contention as to the effect of the words quoted from section 29. The taxpayer presumably would not make two returns of income for its fiscal year *1211917, one for the purposes of the 2 per cent tax and another for the purposes of the 4 per cent tax, but it made a return of its income for that fiscal year, upon which income there was to be assessed the tax to which under the law it was subject. This whole income, after deducting the proper credit, was subject to the 2 per cent tax, but by specific pro* vision of the act of 1917 levying the 4 per cent additional tax, but five-twelfths of that income was subject to the 4 per cent tax, and in so far as the language of section 29 is concerned in its relation to the 4 per cent tax it must be construed as referring to the portion of the net income embraced in the return which under the law is subject to that tax; in other words, by a substitution, the net taxable income embraced in the return.
But let us assume that there is no basis for the construction based upon the reason stated, or any similar conclusion, and that the case presents itself as one requiring the construction of a statute and the adoption of one of two conflicting theories where the statutes themselves do not furnish the basis for a conclusion. We then have, with reference to these two methods, a situation, exemplified by illustrations in plaintiff’s brief and which we understand are not in dispute, involving radically different results in this, that the method adopted by the Bureau of Internal Revenue results in inequalities, while the method contended for by plaintiff eliminates these inequalities. It is demonstrated by illustrative calculations that the bureau’s method not only results in inequality as between calendar year and fiscal year corporations but necessarily results in inequality as between fiscal year corporations dependent upon the number of months in the fiscal year which are included in the calendar year. We need not quote the examples given. They demonstrate this result beyond peradventure of doubt. Because of this situation the conclusion necessarily to be reached has been determined for us by the Supreme Court.
In the case of Knowlton v. Moore, 178 U. S. 41, in which the then Chief Justice wrote an elaborate and able opinion frequently referred to, the condition of inequality here apparent as between two different methods was also present. *122In discussing that feature of the case the writer of the opinion illustrated by his own method this condition. He refers to the fact that there, as here, tables are found in counsel’s brief, “ Which show how inevitable and profound are the inequalities which the construction must produce,” to which he adds that “ Clear as is the demonstration which they make they only serve to multiply instances afforded by the one example we have just given,” and he than says “We are, therefore, bound to give heed to the rule, that where a particular construction of a statute will occasion great inconvenience or produce inequality or injustice, that view is to be avoided if another and more reasonable interpretation is present in the statute.”
In Semple & Company v. Lewellyn, 1 Fed. (2d series) 145, the same question was before the District Court of the Western District of Pennsylvania, and in a rather elaborate opinion it was held that the method of computation used by the Bureau of Internal Revenue was erroneous and that the method here contended for by the plaintiff was the correct method. We do not attempt to quote from the opinion because brief quotations would not properly present its full scope, but refer rather to the opinion itself. It is to be added that upon appeal the Circuit Court of Appeals for Third Circuit, at its October term, 1925, affirmed the judgment of the district court. It quoted from the opinion of the district judge, and concluded by saying:
“ Narrowing, as the case does, to following one of the alternative computations, nothing further need be said but that we are in accord with the calculation adopted by the court below and therefore affirm its judgment.”
For the reasons stated we must conclude that the method adopted by the Bureau of Internal Revenue in the assessing of the 4 per cent tax here involved was erroneous and that therefore the plaintiff is entitled to recover $16,087.65, with interest thereon from November 21, 1919, to this date, amounting to $6,188.37, in all the sum of $22,276.02, and we have so ordered.
Geaham, Judge; Hat, Judge; Booth, Judge; and Campbell, Chief Justice, concur.