EDWIN J. HUNTER, Jr., District Judge:

This case involves claims for refund of federal income taxes in excess of $10,000,000. The tax years involved are 1968 and 1969. Following payment of the deficiencies under protest and the disallowance of its claims for refund, plaintiff filed this suit in the District Court. There are three separate and important issues involving interpretations of Internal Revenue Code sections dealing with: (1) employer deductions to qualified employee pension trusts; (2) credit for foreign taxes paid by a Western Hemisphere Trade Corporation; and (3) the distinction between "tangible" and "intangible" property as applied to investment tax credit and the double declining balance method of depreciation. The District Court concluded that Texas Instruments was entitled to the deductions made to its employee pension trust and that it was entitled to carry over the foreign tax credits under Section 1503(b) of the Code, but that it was not entitled to a tax credit and accelerated depreciation for certain tapes and films. We affirm as to the first two issues and reverse as to the latter.

CONTRIBUTIONS TO PENSION TRUST

The TI Employee's Pension Trust ("Pension Trust") is a qualified pension trust organized for the exclusive benefit of the employees of TI and certain of its subsidiaries. Under Section 404(a) of the Internal Revenue Code, employer contributions to the trust are deductible when paid if they are computed pursuant to one of several well-established actuarial methods, and also meet the "ordinary and necessary" business expense test incorporated by reference from Code section 162(a).

On this appeal the Government does not challenge the fact that taxpayer's payments were in accord with an appropriate Code actuarial method, but insists that in the years of payment, the contributions did not satisfy the "ordinary and necessary" business expense standard. They argue that they were not "ordinary" expenses because they were nondeductible capital outlays and alternatively that they were not "ordinary" or "necessary" within the more general meaning of those terms. These contentions will be considered separately. The "ordinary and necessary" business expense test is conjunctive and must be applied with regard to the facts of each case. Welch v. Helvering, 290 U.S. 111, 54 S.Ct. 8, 78 L.Ed. 212 (1933).

During 1968 and 1969, approximately 23,000 employees of TI were covered by the Pension Trust. The primary investment and management responsibility rests with a retirement committee composed of not less than five (5) participants. All necessary actuarial computations are made by or under the supervision of an actuary. Significant uncertainties concerning future events are of necessity constructed into the basic assumptions used to value the contributions. Once the various significant variables are identified and reduced to statistically probable values, the present value of future benefits can be computed. There are several acceptable actuarial methods. This case involves two of the most commonly used — the aggregate method and the entry age normal method. The sub-paragraphs of IRC section 404(a)(1) permit their utilization. Sub-paragraphs (A) and (B) together allow use of the aggregate method; subparagraph (C) allows use of the entry age normal.

For many years prior to 1967, the amount of the contribution made to the Pension Trust was actuarially determined under the entry age normal method. For 1967, the amount of the contribution was determined under the aggregate method. Likewise, the amounts contributed for 1968 and 1969 were determined under the aggregate.

The objective of the aggregate method is to fund the total estimated remaining liabilities over the remaining working life of the participants. The present value of benefits to be paid in the future is determined upon certain assumptions concerning interest rate, administration expenses and future benefits. The present value of future benefits is then reduced by the value of assets to determine the present value of the benefits which have not been funded. The contribution for the year is then equal to a level percentage of payroll which would completely fund the unfunded benefits, if paid over the remaining working life of the plan participants.

Under the entry age normal method, the "normal cost" for the particular year is first determined. These costs are computed on the assumption, (i) that every employee entered the plan (entry age) at the time of employment or at the earliest time he would have been eligible if the plan had been in existence, and (ii) that contributions of "normal cost" have been made from the entry age to the date of actual valuation. Theoretically, the contributions are then equal to the level percentage of payroll which, if contributed each year and accumulated at the rate of interest used in the actuarial valuation, would result in a fund equal to the present value of the pensions at retirement for the employees who survive to that time.

Since "normal cost" under the entry age normal method is determined on the assumption that there have been prior contributions of "normal cost," which may or may not be true, it is necessary to compute separately the past service costs on that group of employees who did not enter the program at its creation. Hence the annual contribution under the entry age normal method is comprised of two components: the normal cost and unfunded past services cost. The past services cost may be reduced by contributions equivalent to the interest on the unfunded balance of past services cost and an amount intended to reduce the principal amount of the unfunded balance.

During 1967 it was deemed advisable to change the interest rate assumption for the Pension Trust from 3 1/2% to 4%. This change was warranted by past investment experience and a forecast of investment probabilities in the future. The interest rate assumption is an important factor in determining the present value of future benefits, and the increase substantially decreased the present value of the benefits payable by the Pension Trust as of January 1, 1967.

At the time the decision was made to increase the interest rate assumption, the actuary for the Pension Trust advised that under the Internal Revenue Service's ("IRS") position, TI would have no deductible contribution to the Pension Trust for 1967 if the contribution were computed under the entry age normal method. Therefore, the actuary recommended that TI change to the aggregate method of computing its contribution to the Pension Trust for 1967.

Some companies which were not in good shape financially had changed their actuarial assumptions in order to reduce or avoid entirely contributions to their pension trusts. TI, as well as the financial community, felt this type of action by these companies did not satisfy established liabilities and, in effect, resulted in improper financial reporting to the public. TI recognized that such action had been received very poorly in the financial community and TI had no desire to be placed in a similar position, either with the financial community or with its own employees. In the words of George Livings, a member of the retirement committee, the Pension trust changed to the aggregate method to "avoid these wild swings where you go from eight million dollars' contribution in one year to zero in the next."

The entry age normal method lends itself to the concept of "overfunding." If the assets exceed accrued liability, the trust is said to be overfunded. The concept of overfunding is alien to the aggregate method, because "accrued" costs of past service are spread, along with current costs, over the remaining work life of each employee. A pension trust under the aggregate method is not strictly "overfunded" until the value of the trust assets exceeds the total benefits payable.

In the light of what has been said, we turn to the precise contention that the contributions were not "ordinary," because "they were non-deductible capital outlays." Shorn of unnecessary actuarial embellishment, the bottom line of this contention is simple. The government puts it this way:

"The taxpayers' Pension Trust was substantially overfunded in 1968, its $7,000,000 contributions in 1968 and 1969 constitute payments for (or attributable to) its employees' pensions for their future years of service."

We do not regard as contrarily persuasive the numerous decisions adhering to the established rule that prepaid expenses are not deductible in the year of payment and must be capitalized because they create an asset having a longer life than a single tax year. E. g., Commissioner of Internal Revenue v. Lincoln Savings Loan Association, 403 U.S. 345, 91 S.Ct. 1893, 29 L.Ed.2d 519 (1971); Burck v. Commissioner of Internal Revenue, 533 F.2d 768 (2nd Cir. 1976); Commissioner of Internal Revenue v. Boylston Market Association, 131 F.2d 966 (1st Cir. 1942). No authority is cited and we have found none supporting the contention that contributions to pension trusts must be capitalized because the value of its assets exceeds its "accrued liabilities" computed under the entry age normal method.

Generally, an expense is considered prepaid if it is paid in a taxable year prior to the taxable year in which the benefits therefrom are received. The non-deductibility of a prepaid expense is based on the assumption that an expense in a fixed amount can be directly associated with a benefit to be received in a future year. Paying a worker so many dollars per hour is a simple closed-end business transaction. Contrast this with a plan to pay a 21 year old worker a pension when he retires. Differences in life expectancies, salaries' earning rates, inflation and a host of other factors doom simplistic answers to uselessness. In the pension funding area, many present benefits cannot be objectively gauged, all benefits will be received by employees in the future, and the calculation of the annual pension contribution is determined by what actuarial computation is used.

Every case which has considered the meaning of the phrase "ordinary and necessary" in connection with contributions to pension trusts has held that payments to pension trusts are deductible if the total amount of compensation paid, including pension contributions, does not exceed reasonable compensation for personal services actually rendered.

The government insists that the value of the Pension Trust's assets exceeds its "accrued pension liabilities." The phrase "accrued pension liabilities" as used by IRS refers to past service costs and normal costs computed under the entry age normal method. The effect of this capitalization argument is to impose the entry age normal method on TI, and other taxpayers, despite the fact that 404(a)(1) and the regulations clearly recognized that the aggregate method is a reasonable actuary method for computing the amount of contributions to a pension trust. The district court properly determined that TI's 1968 and 1969 contributions were ordinary expenses and did not constitute non-deductible capital investments.

We turn next to the government's position that the contributions were neither necessary nor ordinary in the more general sense. Many cases deal with the problem. To attempt to harmonize them would be a futile task. In Tulia Feedlot, Inc. v. United States, 513 F.2d 800 (5th Cir. 1975), this court had occasion to set forth its view in terms which are particularly applicable:

Expenses must be necessary in the sense that they are, at least, appropriate and helpful for the development of the taxpayer's business. Commissioner of Internal Revenue v. Tellier, 1966, 383 U.S. 687, 689, 86 S.Ct. 1118, 16 L.Ed.2d 185. Section 162(a) does not require that expenses be necessary in a philosophic or logical sense. As one commentator has suggested, a reading of Section 162(a) based on the logical meaning of "necessary" would "place the courts and the Commissioner in the position of business efficiency experts reviewing the commercial decisions of the taxpayer, a function which they are ill-fitted to perform." Comment, Business Expenses, Disallowance And Public Policy: Some Problems of Sanctioning With The Internal Revenue Code, 72 Yale L.J. 108, 113, n. 20 (1962). Because courts are reluctant to review commercial decisions made by a taxpayer, they have generally looked to the actual practices of businessmen to determine the standard. In determining whether expenses are ordinary and necessary under Section 162, the test is whether a hard-headed businessman, under the circumstances would have incurred the expense. Cole v. Commissioner of Internal Revenue, 2 Cir. 1973, 481 F.2d 872, 876. This rule is, of course, subject to numerous limitations. First, it is only the business practices of a hard-headed businessman, and not his tax avoidance schemes, that are subject to judicial deference.

The very premise upon which the government argument is pegged — that TI changed to the aggregate method solely for tax avoidance purposes — directly conflicts with the findings of the district court:

* * * but the evidence in this case demonstrated that TI was carefully attempting to hew a path through the uncertain future using computers, surveys, regression analysis and past experience as its cutting edge. While blunt, the tools used by TI are the only ones available to mortal man to predict the inscrutable future. In the instant case, it is clear that TI contributed monies to its pension fund, not in blind adherence to an arithmetic scheme, but out of a well reasoned faith that the average figures chosen as a basis for its contributions would prove accurate in the long run.

* * * * * *

I conclude therefore that the contributions in this case were ordinary and necessary within the meaning of Section 404(a) because they were based upon assumptions which were honestly, conscientiously, and scientifically derived and because they were in good faith applied in 1968 and 1969. 407 F. Supp. 1326 at 1335.

No useful purpose will be served in further detailing the extensive evidence. It suffices to say that the district court's decision is based upon substantial evidence, that it was not induced by any erroneous view of the law, and that it is not erroneous. The 1968 and 1969 contributions are deductible as necessary and ordinary expenses.

FOREIGN TAX CREDIT CARRYOVER

As a United States citizen, TI is taxed on its income form all sources, whether domestic or foreign. In order to prevent double taxation on that income which is taxed by a foreign country the taxpayer is allowed to take a credit against its United States taxes by the amount of the tax paid to the foreign country.

A number of rules limit the amount of this foreign tax credit. A basic general rule is that a taxpayer cannot take a credit in an amount larger than that portion of its United States tax applicable to the income on which the foreign tax was paid. This basic limitation on foreign tax credits may be computed in two ways. One method is referred to as the per-country limitation. The other is the overall limitation. If a taxpayer elects the per-country limitation, the limitation is computed separately as to the taxes on income in each separate foreign country. If a taxpayer elects the overall limitation it, in effect, combines all foreign sourced income and foreign tax credits and determines the limitation on a combined basis.

GSIF is a Western Hemisphere trade corporation (WHTC). In general, a WHTC is a corporation which does all of its business in the Western Hemisphere, derives 90% of its business from outside the United States, and is actively engaged in business. IRC 921.

A special rule applies where a WHTC joins in the filing of a consolidated return. Section 1503(b) provides:

"(1) If the affiliated group includes one or more Western Hemisphere trade corporations (as defined in § 921), and if for the taxable year an election under § 904(b)(1) (relating to election of overall limitation on foreign tax credit) is in effect, then the amount of taxes paid or accrued to foreign countries and possessions of the United States by such Western Hemisphere trade corporations which may be taken into account for purposes of § 901 shall be reduced by the amount (if any) by which —

"(A) the amount of such taxes * * * exceeds

"(B) the amount of the tax computed under subsection (a) [relating to the consolidated return regulations] with respect to the portion of the consolidated taxable income attributable to such corporations." (emphasis added).

Neither party disputes the fact that the motivation and impact of § 1503(b) is to segregate the foreign tax credit computations of a WHTC within a consolidated group to the extent a WHTC pays lower U.S. taxes than other members of the consolidated group and, thus, prevent the use of that differential during the taxable year to reduce the taxes of the other members of the group.

In 1968 and 1969, GSIF, and other members of the TI consolidated group, paid foreign taxes and elected the overall limitation method under § 904(b)(2). The foreign taxes paid by GSIF exceeded the tax on the consolidated taxable income attributable to GSIF by $337,170. Pursuant to § 1503(b), the amount of foreign tax credits to which TI was otherwise entitled in 1968 was reduced by $337,170. Taxpayer argues that it may carry this 1968 credit into 1969 and reduce its taxes for 1969 by approximately that amount.

The issue is: Are the foreign tax credits of $337,170, when reduced under § 1503(b), completely lost, as the government contends, or can they be carried back and forward as other unused foreign tax credits? The answer must be determined from statutory interpretation and legislative history since the regulations are silent and no cases have considered this issue.

No provision in the Internal Revenue Code or the regulations expressly prohibits the carry over or carry back of the foreign tax credit under § 1503(b).

The legislative history of § 1503(b) supports the conclusion that Congress never intended that the reduction under § 1503(b) should produce a total loss of the reduced credits. Section 1503(b) resulted from a Senate amendment to the House version of Pub.L. No. 86-780, 86th Cong., 2d Sess. (Sept. 14, 1960), the same law under which § 904(a)(2) (the overall limitation provision) was added to the Code. The Senate Finance Committee explained the amendment as follows:

"Under the House bill where a consolidated return is filed, foreign taxes which cannot be credited against U.S. tax [because they exceed the 38% rate in effect for Western Hemisphere trade corporations as of the time of the Finance Committee's report], can be used to offset U.S. taxes at 52% [the ordinary corporate then in effect] on other foreign-source income where the foreign tax rates involved are not this high. For example, assume that the income of a Western Hemisphere trade corporation is $100 before the imposition of foreign taxes of $45. Assume another domestic corporation also earns $100 in another foreign country and is subject to the same $45 foreign tax, but that this company is not a Western Hemisphere trade corporation. The Western Hemisphere trade corporation in this case would generally be subject to a U.S. tax (before foreign tax credit) of about $38 (ignoring the surtax exemption). Therefore, in this case only $38 of the $45 of foreign taxes could be credited against U.S. tax, leaving $7 of foreign taxes which cannot be credited. In the case of the other corporation, the U.S. tax before foreign tax credit would be $52 (again ignoring the surtax exemption). Against this could be credited the full $45 tax paid to the foreign country, leaving a net U.S. tax of $7. If the income of the two corporations in this example were included in a consolidated return, it would be possible in effect to credit the $7 of foreign tax not credited in the case of the Western Hemisphere trade corporation against the $7 of U.S. tax otherwise due in the case of the corporation subject to the 52% tax. To prevent this result, the amendments made by your committee provide the foreign taxes which cannot be credited against U.S. taxes in the case of Western Hemisphere trade corporations as a result of the special 14-point-tax differential provided for these corporations may not be used to offset U.S. tax on other foreign-source income either in the current year or in the years to which the unused credits may be carried. This is applied only where a consolidated return is filed and only where the overall limitation is used." S.Rep. No. 1393, 86th Cong., 2d Sess. pp. 3770, 3775 (1960), 1960-2 Cum.Bull. 874, at 878. (emphasis added).

The underscored language clearly reveals that Congress intended that the foreign tax credits reduced pursuant to § 1503(b) were subject to carry back and carry over provisions. Nowhere in the legislative history is there any suggestion to the contrary. The reference to "unused credits" could only refer to those reduced. The statement is that these reduced credits cannot be used in the taxable year, or in the years to which they may be carried, to offset U.S. tax on other foreign-sourced income.

Such an interpretation of § 1503(b) and § 904(d) is not awkward as the government asserts. The foreign tax credits reduced under § 1503(b)(1) in 1968 may be carried forward to 1969 under 904(d). The credit for the remaining corporations, however, would be limited by § 904 alone and would be determined as though they were the only members of the consolidated group. Thus, the carry over of WHTC foreign tax credits would not be used to offset foreign tax credits attributable to non-WHTC corporations. This is the application of § 1503(b)(1) clearly envisioned by the Senate Finance Committee report at the time of enactment of § 1503(b).

Acceptance of the government's position would undermine statutory intent and cause a result which treats a WHTC filing a consolidated return differently than if a separate return were filed. Acceptance of the government's position results in a complete loss of foreign tax credits in the only situation in which foreign tax credits are limited or reduced in a taxable year without benefit of the carry back and carry forward provisions. Nowhere in the legislative history is there to be found any implication that Congress sought such a far-reaching result.

The conclusion is inescapable. Texas Instruments, Inc. is entitled to carry forward losses sustained by its foreign subsidiary GSIF in 1968, pursuant to Section 904(d).

INVESTMENT TAX CREDIT AND DEPRECIATION

A taxpayer is entitled to claim an investment tax credit equal to a specified percent of the basis of "section 38 property." Under § 48(a)(1), the term "section 38 property" is defined to include "tangible personal property." Similarly, a taxpayer is entitled to a deduction for depreciation under the double-declining balance method, if the depreciable property is "tangible personal property."

During 1968 and 1969 plaintiff's subsidiary, Geophysical Service, Inc. (GSID), was engaged in the business of collecting, processing and selling or licensing offshore seismic information to various customers who in turn used that information to explore for oil and gas. While the information was furnished to the customer in picture from depicting the contours of the earth's different strata, the actual collection and editing process involved a complicated computer process.

The method used in collecting the seismic data needed to produce such pictures was to introduce sound into the ground and then capture the various reflected vibrations from the subterrain in microphone-like receivers. Those receivers then transmitted the electronic impulses to recording stations where the impulses were transcribed onto magnetic computer tapes known as "field" tapes. From there the impulses recorded on the field tapes were taken to a processing center where background noise or signals were eliminated. With the retained or primary signals sharpened by the editing process, a "final" or "output" tape was produced. Using a computer, the information contained on the output tapes as electronic impulses was then transformed into a picture representing a vertical slice of the earth. The computers through which the field tapes were processed are digital computers and the reflex signal data were placed on the output tapes in digital form.

When a customer placed an order for the information collected and processed by GSID, he received a copy of the original picture produced by the process, a map locating the points where the sound waves were introduced into the earth, and a report outlining the conditions under which the tests were conducted. GSID retained all field and output tapes as well as the original analog film, since the original picture might be damaged and need to be reproduced or the editing technology might advance beyond the then present state of the art, thereby making it possible to produce more accurate representations of the subterrain from the original recordings. Information furnished on the picture to customers was licensed on a non-exclusive basis pursuant to written agreement. Generally, the customer was restricted from making the data available to any person except those specifically exempted by agreement.

While only two or three percent of the total sales volume of GSID and its related companies was derived from the sale or licensing of the seismic pictures in 1968 and 1969, the costs incurred were substantial. In 1968 the costs were $934,070 and in 1969 they were $2,243,973. On the consolidated return for those years, taxpayer deducted these expenditures as ordinary and necessary business expenses. IRS determined that the cost of this speculative data should be capitalized and then amortized over a seven-year period with a ten percent residual value. TI did not dispute the Commissioner's determination that these costs should be capitalized rather than expensed, but insists that it is entitled to an investment tax credit and the use of the double declining balance method of depreciation on the total capitalized costs of the field tapes, output tapes, and analog film. IRS contends, however, that these tax benefits are applicable only to the cost of the raw tape and film itself, not to the full cost of producing the tapes and film with seismic data thereon. Both parties agree that resolution of the issue turns on whether the tapes and film with recorded seismic information are "tangible" or "intangible" personal property. This is so because the benefits of investment tax credit and double declining balance method of depreciation are applicable only to "tangible personal property." 26 U.S.C. §§ 48(a), 167(c).

The district court held for the government on alternative grounds. 407 F. Supp. at 1326. First, the court held that when a taxpayer puts into service tangible personal property he produced himself, the investment tax credit is available only as to the tangible inputs used. Thus, labor and other intangible costs would be excluded. The court then concluded that the burden was on the taxpayer to propose an allocation between tangible and intangible inputs and that since TI failed to allocate its costs, it could not claim the investment tax credit. Id. at 1341. Second, the court held that TI in incurring costs of producing and processing the seismic data on the tapes and film was not in reality making an investment in tangible property but in intangible information. Id. at 1342.

The government seeks to sustain the district court judgment on the second ground, admitting that the court's analysis on the first ground was erroneous. The government's position on appeal is that if the capital asset in which the taxpayer's costs are invested is essentially intangible, then all costs of acquiring or producing that asset constitute the basis of an intangible asset and the investment tax credit, as well as double declining balance depreciation, is unavailable. The issue as stated by TI is whether these tapes and films constitute tangible personal property. On the other hand, the issue as posed by the government is whether the cost of seismic data constituted an investment in intangible information. Our perception of this argument is that property is intangible if its intrinsic value is attributable to its intangible elements rather than to any of its specific tangible embodiments.

Applying the tangibility vs. intangibility test is made difficult by Congress' failure to define either term in the Code. The government urges deference to the Commissioner's interpretation of the words because in IRC section 38(b) Congress specifically authorized the Secretary of the Treasury to prescribe "such regulations as may be necessary to carry out the purpose of [the Investment Tax Credit]." The government focuses on one such regulation, which reads:

§ 1.48-1 Definition of section 38 property.

* * * * * *

(f) Intangible property. Intangible property, such as patents, copyrights, and subscription lists, does not qualify as section 38 property. The cost of intangible property, in the case of a patent or copyright, includes all costs of purchasing or producing the item patented or copyrighted. Thus, in the case of a motion picture or television film or tape, the cost of the intangible property includes manuscript and screenplay costs, the cost of wardrobe and set design, the salaries of cameramen, actors, directors, etc., and all other costs properly includible in the basis of such film or tape. In the case of a book, the cost of the intangible property includes all costs of producing the original copyrighted manuscript, including the cost of illustration, research, and clerical and stenographic help. * * *

Ordinarily, treasury regulations are entitled to considerable weight in construing the statutory language. Brewster v. Gage, 280 U.S. 327, 336, 50 S.Ct. 115, 117, 74 L.Ed. 457, 462 (1930); Lucas v. American Code Co., 280 U.S. 445, 449, 50 S.Ct. 202, 203, 74 L.Ed. 538, 540 (1930); Kramertown Co., Inc. v. Commissioner of Internal Revenue, 488 F.2d 728, 730 (5th Cir. 1974). Here such deference is unwarranted. In 1971 a federal district court held that under the 1962 investment tax credit scheme, which is in issue here, master film negatives used in the motion picture industry constitute tangible personal property under the statute despite the treasury regulation. Walt Disney Productions v. United States, 327 F. Supp. 189 (C.D.Cal. 1971). When the investment tax credit was reenacted in 1971, Congress expressly indicated its agreement with this holding. Although the government argues that references to Walt Disney in the legislative history were only in the context of prospective changes in the law, the Senate Finance Committee was unequivocal:

A court case decided the question [of tangibility of motion picture film] in favor of the taxpayer. The committee agrees with the court that motion picture and TV films are tangible personal property, eligible for the investment credit. S.Rep. No. 92-437, 92d Cong., 1st Sess. 34, 1971 U.S. Code Cong. and Adm.News, pp. 1918, 1941 (1971).

On appeal the Ninth Circuit affirmed the district court decision. Walt Disney Productions v. United States, 480 F.2d 66 (9th Cir. 1973), cert. denied, 415 U.S. 934, 94 S.Ct. 1451, 39 L.Ed.2d 493 (1974). Perceiving the illogic of arguing otherwise, the government in effect admits that TI's films and tapes are analogous to the films at issue in Walt Disney. Yet it argues that the 1971 legislative history is irrelevant to pre-1971 tax years. We agree with the Ninth Circuit's rejection of this argument:

The government urges that this legislative history is irrelevant because the 1971 Act changed prior law, or at least that the history is entitled to little weight because "the views of a subsequent Congress form a hazardous basis for inferring the intent of an earlier one." United States v. Price, 361 U.S. 304, 313, 80 S.Ct. 326, 332, 4 L.Ed.2d 334 (1960). However, although the 1971 re-enactment of the investment credit did change prior law in several ways, see, e. g., Revenue Act of 1971, Pub.L. No. 92-178, § 104(c), 85 Stat. 497, 501, it did not change in the phrase, "tangible personal property" as used in the 1962 Act, Int.Rev. Code of 1954, § 48(a)(1)(A). And, while subsequent legislative history normally is not of controlling weight, it should not be ignored when it is clearly relevant.

[citations omitted]

In this case we give subsequent legislative history special weight, because the inferences flow not from Congressional action or inaction on amendatory legislation, but from explicit Congressional statements of the meaning of a phrase which was unchanged during the period in question. In addition, Representative Mills was Chairman of the House Committee on Ways and Means both in 1962 and in 1971, and many members of the Senate Finance Committee served in both years. The statements by the Finance Committee and by Chairman Mills demonstrate that the Commissioner's regulation was contrary to the Congressional intent. We hold it invalid insofar as it denied an investment credit on the films in question. Id. at 68-69.

The latest decision involving Walt Disney Productions, decided August 5, 1976, also represents a defeat for the government on this issue. Walt Disney Productions v. United States, 549 F.2d 576 (9th Cir. 1976) ("Disney III"). The decision in Disney III completely upheld the Ninth Circuit's prior decision in Disney I with respect to the right of Disney to claim an investment tax credit on the entire basis of its motion picture films. Likewise, the court in Disney III recognized that the district court's decision in this case is not compatible with the decision in Disney I, but expressly approved its prior decision in Disney I. 76-2 U.S.T.C. at 84, 931.

The District Court assumed, we feel mistakenly, that the resultant property in which taxpayer was investing was intangible information. This assumption is based on the premise that the seismic information is the "property" used by GSID in its business. This premise is incorrect. GSID uses only the data tapes and films to produce copies to sell to its customers. Its customers purchase the seismic data tapes and films to obtain the seismic information. Stated differently, taxpayer takes a picture of the subsurface formations of the earth, processes the film into a reproductive negative, and sells prints of that picture to its customers.

The government's arguments, plausible as they may sound, simply refuse to recognize that the value of the seismic data is entirely dependent upon existence of the tapes and film. If the tapes and film were destroyed prior to any reproduction of the film analog, nothing would remain. An investment in the data simply does not exist without recording of the data on tangible property. Thus the basis of the tangible tapes and films must include the costs of collecting seismic tapes and films must include the costs of collecting seismic data and recording it on the tangible property, with the result being an asset constituting "tangible personal property."

The statute admits of no other construction on the facts of this case. The ordinary and customary meaning of "tangible" is that which can be felt by touch, having actual form and substance. Webster's New Twentieth Century Dictionary 1863 (unabridged 2d ed. 1964). The seismic data tapes and films are admitted by the government to be tangible. They have intrinsic value because the seismic information thereon does not exist as property separate from the physical manifestation. The seismic data tapes and films are tangible personal property used by GSID in its business of producing copies for sale. The entire cost of producing such tapes and films must be included in the basis upon which investment tax credit should be allowed.

CONCLUSION

For the reasons stated, the judgment of the district court (i) that the contributions made by TI to the Pension Trust in 1968 and 1969 are deductible as "ordinary and necessary" business expenses in those years under § 404(a); and (ii) that GSIF was entitled in 1969 to a carryover of 1968 foreign tax credits which had been reduced by § 1503(b) are affirmed. Further, the judgment of the district court (i) that certain seismic data tapes and films used by GSID in its business were not section 38 property; and (ii) that certain seismic data tapes and films used by GSID in its business were not depreciable under the double-declining balance method under § 167 is reversed.

AFFIRMED IN PART; REVERSED IN PART and REMANDED.