*604OPINION*
AkuNdell :Wo have no doubt that the individual business of B. F. Sturtevant had a valuable good will at the time of its transfer to petitioner. But where such an intangible asset is acquired, as here conceded, without a payment of stock or other consideration, may all or any part of its value be included in invested capital? The question was fully considered in Herald-Despatch Co., 4 B. T. A. 1096, and the conclusion reached that good will acquired under such circumstances may not be included in invested capital as a paid-in surplus. There has been no departure from this rule. Stephens-Adamson Manufacturing Co., 16 B. T. A. 41, and Concrete Engineering Co., 19 B. T. A. 212. The courts have adopted the same principle. *605Daily Pantagraph v. United States, 37 Fed. (2d) 783; 68 Ct. Cls. 251; Colorado Continental Lumber Co. v. United States, 42 Fed. (2d) 33; 70 Ct. Cls. 413; Baker & Taylor Co. v. United States, 26 Fed. (2d) 187; certiorari denied, 278 U. S. 615; LaFayette-South Side Bank v. Commissioner, 33 Fed. (2d) 646, affirming 7 B. T. A. 1307. This issue is decided in favor of respondent.
The major issue involves the value at March 1,1913, for exhaustion purposes of patents and licenses to use patents employed by petitioner in the taxable years in the production of some of its products. Considerable evidence was submitted in the form of testimony, sales charts, earnings, etc., and many facts, including the remaining life of the patents at the basic date, were stipulated.
The parties differ considerably in their contentions as to the value of the patents. Of the fifty-nine patents in' force and use by petitioner on March 1, 1913, sixteen, divided into four groups classified as multivane fan patents, fuel economizer patents, slow-speed fan licenses, and turbine patents, are the subject of valuation. The petitioner is contending for a valuation of $1,175,000 for the four groups of patents. The respondent has allowed depreciation on a value of about $100,000 based upon costs of purchase and development, and asks that his allowance be sustained. In support of its claimed valuations petitioner introduced opinion evidence and a great mass of statistical data, the latter being designed to show the earnings attributable to patented articles.
In the absence of actual sales of property, evidence of this sort is of aid in determining what a willing buyer might reasonably be expected to pay. We do not understand, however, that either opinion evidence or a calculation based on conjectural earnings are conclusively determinative of value. Tracy v. Commissioner, 53 Fed. (2d) 575; Uncasville Manufacturing Co. v. Commissioner, 55 Fed. (2d) 893; Reinecke v. Spalding, 280 U. S. 227. It is our right to determine the weight to be given to the evidence upon which either party relies in the light of other facts developed in the case. Anchor Co. v. Commissioner, 42 Fed. (2d) 99.
The petitioner here is undoubtedly confronted with a difficult case to prove. It manufactured a variety of devices under a number of patents, and in addition it apparently had a substantial good will, so that even if its earnings were larger than would ordinarily be attributable to tangibles, it is no easy matter to segregate the earnings and allocate so much to this or that particular intangible. Petitioner has attempted to segregate the profits on the patented articles and those on the nonpatented articles for the 5%-year period prior to 1913 and to fix a value for the patents by capitalizing as income attributable to the patents the amount by which the profit on each group of patented articles is claimed to exceed the profit which *606would have been derived from an equal volume of gross sales if the percentage of profit to gross sales were no greater than in the case of nonpatented articles. Assuming that this method might be of aid in fixing values under some conditions, in this particular case we are unable to accept the result reached, because of the obvious lack of accuracy in the basic figures. Petitioner’s records for the period before 1913 were not sufficiently detailed to permit of a direct and accurate computation of'the profits on either patented or nonpatented articles, and so it resorted to a circuitous method which was described in this way: The amount of sales of all products was determined. From this figure the amount of sales of patented articles was deducted, giving the sales of nonpatented articles. Next, the cost of all products was set down, from which was subtracted the cost of patented articles and the difference represented cost of nonpatented articles. Then the profit on nonpatented articles was determined by subtracting the cost thereof from the sales, and on these figures the percentage of profit to sales was determined. When this percentage was determined, each group of patents was taken separately, the cost of sales deducted from sales, and from that figure there was subtracted an amount computed by multiplying the sales of each group by the percentage arrived at as above described. The resultant figure in the case of each group of patents is claimed to be the excess profit attributable to the patents. The procedure may perhaps be more readily understood by setting out the figures submitted for one of the several years:
[[Image here]]
*607[[Image here]]
This computation showing how the claimed excess earnings on multivane fans for one year was reached is illustrative of the computation made with respect to each of the groups of patents for each of the fiscal years in the period July 1, 1907, to June 30, 1912, and the last six months of 1912. The method above outlined is called the “ primary method ” by petitioner. A further computation was made on each group of patents by deducting from the profits 8 per cent of the tangible assets used in the manufacture of each article, the amount of the assets so used being determined by applying to the total tangibles the percentage representing the ratio of sales of patented articles to total sales. This method is called the “ secondary method.” The resultant figures under each method were capitalized by applying a formula of 10 per cent for income and 4 per cent for sinking fund. Under each method the result is practically the same. These results, petitioner says, establish the values of the several groups of patents.
The first figure used, sales of all products, is taken from petitioner’s books. This figure, it appears, includes outside erection costs. The amount of such costs is given for only one year, the year ended June 30, 1908, when they amounted to $253,535.91, or something over 12 per cent of the figure listed as total sales for that year. The amount of erection costs was not allocated to both patented and nonpatented articles in the segregation subsequently made, but was all carried into the figure claimed to represent sales of nonpatented articles.
The second item in the computation is called sale of patented articles, i. e., multivane fans, turbines, fuel economizers, and slow-speed fans. These figures were compiled from sales orders in petitioner’s files. The tabulations as submitted to us contain numerous omissions of prices. In these instances, it is explained, no prices were shown on the orders. Some idea of the extent of the omissions may be gained from the following examples. On one page listing twenty-eight sales of multivane fans the sale price is omitted in twelve instances; one sheet of nineteen sales has ten omissions; another listing fifty-seven sales contains no sale price for thirty-eight items. *608There are thirteen sheets listing sales of turbines and of these only fonr are complete. Greater accuracy was possible with sales of fuel economizers. Out of about two hundred sales listed, the selling price is disclosed in all but four instances. There are thirteen sheets listing sales of slow-speed fans and none of these is complete.
It is impossible to state accurately what percentage of total sales is omitted, but from the list prices contained in the tabulations it would appear that the omissions represent substantial portions of total amounts. For example, in the fiscal year 1909 the list prices of items omitted amount to over $49,000, against sales recorded of $179,686.64; for the fiscal year 1912 list prices omitted are over $98,000 and sales recorded amount to $617,288.80.
Moreover, there is such a wide variation of sales prices shown in the case of slow-speed fans — which is the only group susceptible of being readily checked in this respect — that we are reluctant to accept the figures as accurate. For example, the sale prices of the Number 40 fan varies from $58.50 to $150. Between these two extremes are prices of $67.50, $76.50, $90 and $99. In each instance the same amount, $49.55, is listed as cost. The Number 50 fans, according to the sales sheets, were sold at prices ranging all the way from $73.50 to $180, with a cost of $68.27. The following are a few of the various sales prices' given for the Number 70 fans: $31.34, $73.81, $227.50, $276.25, $363, and $552.50. Against each of these a cost of $116.75 is listed.
A possible explanation of the above extreme variations may lie in the fact that in many cases petitioner sold a combination of its products as a unit; for instance, a fan and a turbine to operate it were sold to a customer and installed as one article rather than as component, parts. In at least some of such cases it appears that petitioner recorded one article as sold at a profit and another at a loss. A few examples will suffice to illustrate the method. Under order No. 171690 it sold four multivane fans and four turbines. The selling price of the fans is given as $876 and turbines, $1,377, against which costs of $472.72 and $1,775.80, respectively, are shown. This reflects a gain of $403.28 on the fans and a loss of $390.80 on the turbines. Computing the sale as that of a unit, the actual profit was but $4.48. Order No. 228951, obtained August 9, 1912, was for one multivane fan and one turbine. A profit of $308.87 is shown for -the fan and a loss of $36 for the turbine. A gain of $138.82 is shown under order No. 230200 for the sale of two multivane fans and a loss of $343.50 on the two turbines sold with the fans, resulting in a loss of $204.68 on the two units. It might be argued that this method would not affect the final result when the patents are considered as a whole. The difficulty is that we do not know that *609the practice was confined to patented articles. If patented and non-patented articles were sold as units, an arbitrary allocation as above illustrated might seriously distort the profits on either group. An example of this occurs in the last six months of the year 1912, in which the alleged profits from the several groups of patents are set forth as follows:
[[Image here]]
In this period it is claimed that the net earnings were only $82,619.66, and the difference of $47,111.38 is said to be a loss sustained on non-patented articles. According to petitioner’s figures for this period sales of patented articles aggregated only $390,263.20, as against sales of nonpatented articles of $1,017,847.45. Even conceding that the patented articles gave a somewhat higher margin of profit, it is difficult to believe that there was such a wide variation as petitioner’s figures purport to show. It is also to be remembered that the items omitted from the tabulations would, if included, further increase the alleged profit on patented articles and increase the claimed loss on the nonpatented items. To accept petitioner’s figures, we would have to believe that the greater part of its products were deliberately sold at a loss.
Petitioner had no accurate cost records in the period prior to 1913. The cost figures used in the computation before us were made up from a few figures compiled by the cost department from time to time and forwarded to the general manager. The method of compiling the figures was to add 10 per cent to cost of material to cover handling and waste, 100 per cent to labor cost to cover shop administration, factory overhead, depreciation, and nonproductive labor and supplies, and then add to the sum of these figures 15 per cent to cover selling costs. The witness who prepared the costs testified that he checked the accuracy of the method and that for the 5%-year period prior to 1913 there was only a difference of about $19,000 on a total of about $11,000,000. But no satisfactory explanation was given of how the check was made. He was asked how he applied the check and he replied that he took the materials purchased during the year, added 10 per cent, then took productive labor and added 100 per cent, and added 15 per cent to the total. We think that we are entitled to a better explanantion than this before being asked to accept a result in which the possibility of error is so manifest.
*610It also appears that some of petitioner’s so-called nonpatented articles were not such in fact. In each of the years 1908 to 1913 petitioner owned the following numbers of patents: 1908,21; 1909,25; 1910, 28; 1911, 85; 1912, 46; 1913, 59.
Only one witness, Ernest P. Freeman, was questioned as to the patents other than those sought to be valued here. Freeman has been connected with petitioner since 1903 and has been general manager and vice president since 1909. At first he said he did not know of any other patents, and later he mentioned several others, but said they were not regarded as of importance. Inasmuch as one of the basic factors in petitioner’s computation is the segregation of its products into the two classes, patented and nonpatented, we think it was incumbent on petitioner to furnish at least enough information to permit us to form some idea as to the importance of the other patents. On the meagre evidence on this phase of the case we do not feel that we can accept petitioner’s classification of its products into patented and nonpatented groups.
Under petitioner’s so-called secondary method, as explained above, an attempt was made to allocate to the patented articles an 8 per cent return on the tangibles used in their manufacture and then to capitalize the balance of the earnings from such articles. It does not appear clearly just what was included in the term “ net tangible assets ” upon which the computation is based. From the testimony of the accountant who made the computation we gather that only plant and physical equipment were included and no consideration was given to other assets such as cash, accounts receivable, and inventory. Another objection to the computation is the manner in which the portion of tangibles allocated to patented articles was determined. This was found by applying to the total of the so-called tangibles the percentage that the sales of patented articles bore to total sales. We have no way of determining whether the resultant figures are even approximately correct. Several of petitioner’s witnesses have been connected with it for a long time. Eugene N. Foss, petitioner’s president, has been with it and its predecessor business for fifty years. Ernest P. Freeman, above mentioned, has been associated with petitioner since 1903; Eugene B. Williams, since 1905; Merton S. Leonard, since 1901; and Herbert M. Fish, since 1902. All of these men held important positions with petitioner. It does seem to us that, with the testimony of these witnesses available, it would have been possible for petitioner to have produced more accurate evidence as to the extent assets were used in the manufacture of patented' articles if the patents really played as important a part in the business as we are now asked to believe. It is seldom that a taxpayer comes before us with such an array of wit*611nesses who have grown up with the business as in this ease. Yet with these witnesses available we are asked to accept as a basis for determining values a set of figures which throughout arc compiled in an arbitrary method. On the meagre evidence that we have concerning petitioner’s tangibles, it appears that the 8 per cent return used is too low. Freeman’s testimony is that in a businesss such as that of petitioner’s it is customary to figure a 10 per cent return. F reeman further testified that it was petitioner’s practice to charge off many items that other companies would capitalize, and so even if the figures used by the accountant in making; up the computation before us represented book assets, the actual assets used in the business very probably had a considerably greater value. The method used is also subject to the criticism that in figuring the percentage of tangibles the sales figure used as a basis is inaccurate because of the omissions in the tabulations of patented sales which we referred to above.
We accordingly conclude that the formula worked out by petitioner can not be accepted as establishing values at March 1, 1913.
We might further point out that even if the figures purporting to show earnings in excess of a fair return on tangibles were accurate, we still would not be able to say that the excess should be attributed entirely to patents. Petitioner’s witnesses, Foss, Freeman, and Orrok, were unanimous in their testimony that petitioner had a valuable good will. Several of petitioner’s witnesses expressed opinions of the value of the patents. These, for the ¡most part, are based on calculations above discussed and which, as we have pointed out, contain too many inaccuracies and arbitrary premises to be of aid in fixing any value. For example, Freeman, in giving his opinion on the multivane fan patent, said it was based on a showing of earnings attributable to that patent of over $100,000 a year. He a dmitted that if the “ figures are wrong as regards that point, my testimony is wrong, but I thoroughly believe it is right * * Freeman was further ox the opinion that the petitioner’s net tangible assets amounted to between three and four millions of dollars prior to 1913, on which it ought to receive a return of 10 per cent before attributing anything to intangibles. As petitioner’s average earnings were only a little over $200,000, it is clear that Freeman’s opinion lacks a reliable basis. When asked whether he thought that the patent could have been sold to a prospective purchaser at the figure he named, Freeman said that there were one or two competitors that would have paid the price “ with our figures before them.” As to the other items sought to be valued here, Freeman testified that he fixed their value along the same general line.
*612Petitioner’s other witnesses, Orrok and Williams, who gave opinions on the multivane fan patent, based their conclusions largely on the sales and earnings figures above discussed.
Freeman was the only witness called to express an opinion as to the values of the fuel economizer patents. On direct examination his testimony was that they were worth “ appreciably more than the $178,000 that the formula shows.” Upon being asked on cross-examination as to the value he had expressed, he said that it was $115,000 or $120,000.
When it came to valuing turbine patents petitioner did not rely on its formula applied to the other items, apparently for the reason that the figures made up show losses in all years except the period July 1, 1911, to December 31, 1912, in which profits of only $849.67 •are shown. According to the tabulations in evidence, turbines were made and sold during all of the period from 1907 to 1911, but at a loss. This situation probably may be accounted for by petitioner’s method, as explained above, of arbitrarily assigning large profits to fans where fans and turbines were sold as units, and little or no profit to the turbines. Orrok’s opinion of the value of the turbine patents was based on purely hypothetical factors. He estimated the amount of horse power that might be developed by turbines, and that petitioner might expect to receive a certain portion of the turbine business. On that basis he estimated the amount of profit at the rate of $2 per horse power after allowing for an 8 per cent return on tangibles. The record does not show that Orrok had any idea of the amount of tangibles that petitioner had or might be required to have in order to produce the estimated horse power. Difference in turbines made by two concerns and difference in management might easily result in a wide variation in this factor. Furthermore, the evidence shows that in fixing his value Orrok was not aware of the infringement suit of the Terry Company that was pending on March 1, 1913, and gave it no consideration.
At March 1,1913, petitioner’s nearest comparable competitor in the turbine business was the Terry Steam Turbine Company, which manufactured nothing but turbines. In 1913 that company’s gross sales were $410,000 and net income was $33,000. We do not consider that these facts furnish any aid in determining the question here. If petitioner’s business in turbines was expected to amount to the same volume as that of the Terry Company it would be at some unknown time in the future after it had succeeded in meeting the competition of the several companies which the evidence shows were then in the field, and after it had built up plant and equipment of *613an unknown amount. We are not informed as to the amount of the Terry Company’s tangibles, nor as to whether it earned anything beyond a reasonable return thereon.
Bentley, the inventor of petitioner’s turbines, was of the opinion that the patents were very valuable “ to the Sturtevant Company.” Our problem is to determine fair market value and Bentley’s testimony does not aid in establishing this. Moreover, it is shown that prior to the sale of his patents to the petitioner, Bentley had offered them to other concerns and his best offer was $3,000 from the General Electric Company, which was then in the turbine business and presumably was in a position to know the value of turbine patents. It is difficult to believe that patents purchased for the sum petitioner paid in 1911 would increase so tremendously in value by 1913, as claimed here, when operations under them had shown a profit of only about $800 (using petitioner’s figures) and had resulted in an infringement suit.
Petitioner did not own the three patents relating to slow-speed fans, but manufactured the fans under license from the owner. The license was acquired without cash consideration or any agreement to pay royalties. The only consideration was that petitioner agreed to use on certain types of work a dust collector on which the licensor held the patents and to pay a royalty therefor. The license was terminable by either party on six months notice. There is in evidence a memorandum signed by the licensor stating that it intended the petitioner should have an exclusive license as long as the conditions of previous agreements were performed. This, however, does not abrogate the cancellation provisions of the license agreement. Under these circumstances it would require a very clear showing to establish that the license had any substantial value. The only evidence directed to proof of value consists of figures based on the alleged earnings, which we are unable to accept.
Considering all the evidence, we are of the opinion that petitioner has failed to establish the fair market value at March 1, 1913, of its several patents and the license under which it manufactured the slow-speed fans, and we are unable to find that it is entitled to use as a basis for exhaustion any greater amount than that used by the respondent.
In reaching this conclusion we have not overlooked or failed to consider the testimony of the several witnesses giving the history of the development of the several patented articles. We have given careful consideration to that evidence, but, for reasons above given, we are unable to find that it supports the values claimed or that it gives us any sound basis for fixing any other values.
*614 Claimed Amortization Deduction.
Most of the facts concerning the claimed deduction for amortization of war facilities were stipulated. The facts as stipulated appear somewhat complicated, but as we understand them they are as follows: One of petitioner’s subsidiaries, the Aeroplane Company, in its separate return originally claimed an amortization deduction for the year ended June 30, 1918. Later, on April 1, 1922, petitioner filed consolidated returns on behalf of itself and subsidiaries, for the years ended June 30, 1918 and 1919, in both of which amortization claims were made on account of assets of the Aeroplane Company. The claims made in the consolidated returns were recommended for disallowance by the respondent’s office, whereupon, on March 27, 1924, petitioner withdrew the claims. In February, 1925, petitioner •filed two briefs, one with the respondent and one with the solicitor of internal revenue, claiming amortization deductions which are in controversy here, and which are not related to the claims previously made on account of the Aeroplane Company’s facilities.
The facilities with respect to which claim is now made were installed between October 28, 1917, and March 3, 1919, for the production of articles contributing to the prosecution of the war. It is stipulated that the Navy Department or its agent “ advanced a portion of the cost of approved additions to [petitioner’s] plant and equipment.” The cost of the improvements was $137,4-08.54. After completion in 1919 of petitioner’s war time contract with the Navy Department, it paid the United States $77,000 and received a bill of sale for the improvements. The $77,000 paid by petitioner represented the value of the improvements in the light of the use to which petitioner then expected to put them, namely, the manufacture of additional blowers for the Navy Department. Because of the subsequent curtailment of armaments by the United States, the petitioner did not receive the further orders it expected.
The parties argue various questions under this issue, such as the effect of the 1919 bill of sale, and the effect of the withdrawal of the first claims filed. In the view we take these questions are immaterial. Section 234 (a) (8) of the Revenue Act of 1918 provides that in the case of the construction of facilities on or after April 6, 1917, for the production of articles contributing to the prosecution of the war then in progress, there should be allowed a “deduction for the amortization of such part of the cost of such facili-tioes * * * as has been borne by the taxpayer * * It is clear in this case that the $77,000 cost to the petitioner was not the cost of war-time facilities as such, but was rather the cost to it of *615facilities which it proposed to use for manufacturing post-war articles. If petitioner’s theory were sound, it would enable taxpayers to secure amortization deductions for war-time facilities in which they had no interest during the war but which were acquired thereafter, even after peace was formally concluded. It seems to •us that it does not matter that the facilities were erected on petitioner’s land and that technically they may have become a part of the realty, for the fact remains that the petitioner had not borne the cost of such facilities while they were used for war-time purposes. We accordingly hold that petitioner is not entitled to the deduction claimed.
Petitioner’s claim for amortization deductions must be denied on the further ground that it was not timely filed. Section 1209 of the Revenue Act of 1926 provides that deductions for amortization of war facilities may be allowed “ if claim therefor was made before June 15, 1924.” The stipulated facts here show that claim with respect to the facilities here involved was first made on February 4, 1925. The claim that had theretofore been filed and withdrawn related to distinctly different facilities. While we have countenanced amendments to claims, United States Refractories Corporation, 9 B. T. A. 671, and have held that a member of an affiliated group may claim amortization deductions with respect to facilities constructed by another corporation within the group, G. M. Standifer Construction Corporation, 4 B. T. A. 525, 550, none of the decided cases go as far as to say that a claim filed by one company and withdrawn will support a claim filed after the statutory period by another with respect to entirely different facilities.
Decision will ~be entered under Rule 50.