BARNES, Circuit Judge:
On December 30, 1966, the Tax Court of the United States held in two related decisions that deficiencies existed in the federal income tax of Eugene L. Free-land and Vera G. Freeland, husband and wife, for the years 1956-1961, and in that of Margaret C. Lowthian for the years 1956-1960. Eugene L. Freeland is now deceased, and his estate, along with Vera G. Freeland, now petitions this court under 26 U.S.C. § 7482 (1964), seeking review of the Tax Court’s decision relating to them. Margaret C. Lowthian also petitions under the same statute for review of the decision in her case, and the two petitions have been consolidated for argument and decision.
In November 1956 the Freelands and Miss Lowthian (hereinafter referred to as “petitioners”) sold their respective interests in a partnership, the Sam Berger Investment Company, the assets of which comprised approximately 4000 acres of land in San Diego, California. They reported the resulting gain on the installment basis as long-term capital gain, a tax treatment which was at first accepted by the District Director of Internal Revenue. In 1962, however, the District Director approved reopening with respect to 1956. (Subsequent years are here involved because of utilization of the installment basis). On July 29, 1964, the Commissioner of Internal Revenue issued statutory deficiency notices, asserting that the income from the sale [575]*575of petitioners’ partnership interests was ordinary income.1 This position was sustained by the Tax Court, and the correctness of that ruling represents the major issue now before us.
The general statutory rule governing sales of partnership interests is stated in section 741 of the Internal Revenue Code, 26 U.S.C. § 741 (1964):
“In the case of a sale or exchange of an interest in a partnership, gain or loss shall be recognized to the transferor partner. Such gain or loss shall be considered as gain or loss from the sale or exchange of a capital asset, except as otherwise provided in section 751 [26 U.S.C. § 751, as amended (Supp. II 1967)] (relating to unrealized receivables and inventory items which have appreciated substantially in value).”
The exception referred to, which is the basis for the Tax Court’s decision, reads as follows:
Ҥ 751. Unrealized receivables and inventory items.
“(a) Sale or exchange of interest in partnership.
“The amount of any money, or the fair market value of any property, received by a transferor partner in exchange for all or a part of his interest in the partnership attributable to—
“ * * *
“(2) inventory items of the partnership which have appreciated substantially in value, shall be considered as an amount realized from the sale or exchange of property other than a capital asset.
«* * *
“(d) Inventory items which have appreciated substantially in value.
“ * * *
“(2) Inventory items.
“For purposes of this subchapter the term 'inventory items’ means—
“(A) property of the partnership of the kind described in section 1221(1) [26 U.S.C. § 1221(1) (1964)] * * *”
The property described in section 1221(1) includes
“stock in trade of the taxpayer or other property of a kind which would properly be included in the inventory of the taxpayer if on hand at the close of the taxable year, or property held by the taxpayer primarily for sale to customers in the ordinary course of his trade or business * * *. ”
Thus the basic issue here is whether the land owned by the petitioners’ partnership was “held * * * primarily for sale to customers in the ordinary course of [its] trade or business.” The Tax Court found that it was, and it is this finding which petitioners dispute.
The question thus posed is established by prior decisions as one of fact. E. g., Richards v. Commissioner of Internal Revenue, 81 F.2d 369, 370, 106 A.L.R. 249 (9th Cir. 1936). Its determination depends upon a fairly complex series of events — a context fully set out in the Tax Court’s findings of fact. There is no real challenge to those findings insofar as they relate to this background, and we therefore adopt them as our statement of facts:
“Petitioners reside in San Diego County, California, and they filed their respective tax returns for the years in question with the district director of internal revenue at Los Angeles, California.
“At all times material Lowthian was [Eugene L.] Freeland’s secretary and business associate. Her participation in the transactions involved herein was through and with Freeland. * * *
“From 1923 through the time of trial, Freeland has been a full time civil and structural engineer living and working in San Diego County, California. He is licensed to practice by California and certain other states and has been and [576]*576continues to be a member of numerous professional engineering associations and societies. During the years involved herein, he was a senior member of a civil engineering firm and a structural engineering firm.
“Freeland’s practice has included land surveying, the design of municipal improvements, the subdivision of land, and the design of buildings and structures. He rendered professional services in some of the larger real estate developments in the San Diego area and included among his clients some of the area’s most active real estate subdividers and developers. Because of the rapid growth of San Diego County prior to and during the years in question, and because of his excellent reputation, Freeland’s practice has been financially successful. The growth of San Diego County has also resulted in a continuing and substantial appreciation in real estate values.
“Among the many land developers employing Freeland and/or his civil engineering firm (hereafter referred to as the ‘engineering partnership’) were George T. Forbes and Theodore M. Jacobs, doing business as Kensington Heights Company. In the late 1940’s and again in 1951 and 1953, the engineering partnership rendered services to Kensington with respect to the so-called Waring Ranch property in the form of surveying the land, preparing boundary, topographical and master plan maps thereof, and representation before the City of San Diego Planning Commission and other city departments in its efforts to have the land annexed to the City of San Diego so as to make it more salable. Four hundred acres were annexed in 1951 and subsequently sold by Kensington. An additional 1,300 acres were annexed in 1952 or 1953 and sold by Kensington to Bollenbaeher & Kelton, Inc., subdividers and developers. The engineering partnership was then employed by Bollenbaeher & Kelton, Inc., to do the engineering work on the subdivision and development of the 1,300 acres.
“In 1953, efforts were made by Kens-ington to have the City of San Diego annex the remaining 4,500 acres. At the time of the annexation proceedings, the engineering partnership prepared a master plan map of the area, which indicated the lines along which development might feasibly proceed. The City of San Diego annexed the 4,500 acres in December of 1953. The 4,500 acres contained a lake, some canyons, and a mountain rising about 1,100 feet above the surrounding terrain.
“Shortly before the annexation, Kens-ington sought Freeland’s help in finding a buyer for the 4,500 acres. Freeland contacted several clients who were not interested because they thought that development and subdivision of the area was too remote in time.
“In April 1954, Kensington gave Free-land a verbal option on the land. Free-land and one of his developer-clients, Sam Berger, decided to form a partnership to purchase the land. On August 6, 1954, they entered into an agreement to form the partnership, the terms of which were to be reduced to writing in the near future. Under the agreement, Berger was to enter into an escrow in his own name for the purchase of the land from Kensington. Before the close of the escrow, Berger was to nominate the partnership to take title to the land.
“Berger negotiated the terms for the purchase with Kensington. In the course of the discussions, Berger indicated that the whole 4,500 acres would be developed as soon as possible and that he had $20,000,000 in financing available, which was not in fact the case. Berger also told Kensington that the purchase price would be paid off in about three years.
“On August 6, 1954, Berger entered into an escrow agreement with Kensing-ton. The agreement provided for a total consideration of $4,676,666.66 (computed on the basis of $800 per acre plus interest at 3ys percent), payable $100,000 down and the balance in 20 annual installments. The purchase price was to be [577]*577evidenced by an installment promissory note, the first installment of $180,000 to become due on August 6, 1955, with the remaining installments due annually thereafter in increasing annual amounts of $6,000 until the 19th installment in the amount of $288,000. The 20th and final installment ./as to be the remaining balance of $130,666.66. The escrow agreement contained provisions for the release of 150 acres against each annual payment, for the privilege of prepayments with appropriate adjustments for interest, and for the deposit by the purchaser of $200,000 to be disbursed for off-site improvements upon instruction from Kensington.
“The term ‘off-site improvements’ comprehended two things:
(a) the bringing of utilities, viz., water, sewer, gas, electricity, and telephone, from a point distant from the boundaries of the property being developed up to the boundaries of such property; and (b) the installation of a utility, e. g., a water main, on the property being developed, which utility is larger than required to serve such property but which is nevertheless required by the City of San Diego in order to be available to serve further outlying property in the path of future development. With regard to these latter ‘off-site improvements,’ the city generally paid the difference between what the property being developed required and what the city insisted upon for purposes of future property development.
“Freeland provided $70,000 towards the $100,000 down payment. Berger borrowed the other $30,000 with Free-land acting as guarantor of the loan.
“In the process of trying to raise the $200,000 deposit, Freeland contacted the General Petroleum Corporation, seeking a loan. As part of the attempt to borrow the funds, a letter was sent on September 14, 1954, over Freeland’s signature, but without identification as to the capacity in which he was signing, to General Petroleum Corporation’s representative in San Diego. The letter proposed that the corporation lend the necessary funds in exchange for exclusive service station sites within the contemplated 4,500-acre development. The letter also stated that a master plan for the development of the entire 4,500 acres was being completed and that it was proposed to start immediately the development of approximately 2,000 homes. The letter further stated that it was anticipated that a minimum of 4,000 homes would be constructed and occupied within three years and that, after the points involving financing the off-site improvements, completing the master plan, and selecting the location of the first 2,000 homes had been settled, consideration would be given to the sale of acreage to subdividers capable of proceeding with the development in a satisfactory fashion. After due consideration, General Petroleum declined to provide any funds.
“After several unsuccessful attempts to raise the $200,000 from various other sources, Berger’s associates in a development operation consisting of a joint venture of several corporations, known as ‘Country Club Park’ (for which Freeland individually acted as a consultant before various governmental agencies), agreed to put up the $200,000 for the off-site improvements. In return, this group (hereafter referred to as the ‘Barenfeld-Glaser Group’) received a 20 percent interest in the Berger-Freelanc partnership.
“On October 21, 1954, the Lake Murray Development Company, a California corporation (hereinafter referred to as ‘LMDC’), was incorporated for the purpose of subdividing and developing land and constructing homes thereon. LMDC was capitalized with funds that were advanced by the Barenfeld-Glaser Group. The stock of LMDC was held by Herbert Glaser, an attorney, as trustee for the benefit of the corporations which comprised the Country Club Park joint venture. Berger was the initial president of LMDC and Herbert Glaser was its secretary and attorney. Neither Free-land nor Lowthian was an officer, director, stockholder, or investor in LMDC. [578]*578Other principals included Sam Glaser, the father of Herbert, and Jack Baren-feld.
“Shortly after October 21, 1954, but prior to October 26, 1954, Sam Glaser and Jack Barenfeld caused $200,000 to be advanced by the Country Club Park joint venture to LMDC and thence from LMDC to the Berger-Freeland partnership, Sam Berger Investment Company (hereinafter referred to as ‘SBIC’). On October 25,1954, SBIC caused the $200,-000 to be deposited in the escrow covering the purchase of the 4,500 acres. On October 26, 1954, the escrow was closed and title in the land was vested in SBIC subject to a deed of trust for the unpaid purchase price.
“On or about October 26, 1954, SBIC and LMDC entered into an option agreement, which was subsequently reduced to writing and executed on November 23, 1954. Partial consideration for this option was the fact that Sam Glaser and Jack Barenfeld had caused LMDC to provide the $200,000 to SBIC to enable it to close the escrow. The agreement gave LMDC an option to purchase 500 acres of land lying within a larger parcel of approximately 800 acres in the southeast corner of the property for a price of $2,000 per acre. The option agreement further provided in pertinent part:
“(a) That SBIC was to release from the Kensington escrow a minimum number of acres per year in order to prevent a lapse of its right to continue to release acreage.
“(b) That if LMDC purchased less than 150 acres per year under its option, SBIC had the right to sell the unpurchased difference.
“(c) That LMDC realized that SBIC had an oral arrangement with ‘other subdividers’ to purchase portions of the 800-acre parcel within which the 500 acres subject to the option was encompassed.
“(d) That SBIC would cooperate with LMDC in filing the necessary maps and in expeditiously completing the building lots.
“(e) That LMDC would submit to SBIC all its contracts with subcontractors for all items of work so as to guarantee that the cost would not exceed a certain minimum.
“(f) That LMDC should be reimbursed by SBIC for expenses in excess of the stipulated maximum per lot cost of development, and SBIC would retain, as profit, any funds remaining in the trust fund established for lot development.
“(g) That SBIC would be liable for all off-site improvement costs in excess of the $200,000 originally advanced by LMDC.
“On November 23, 1954, the date on which the option agreement between SBIC and LMDC was signed, a formal limited partnership agreement was also executed by all of the general and limited partners in SBIC. This agreement was effective as of August 6, 1954. Under the terms of the agreement, Free-land, with a 32 percent interest, and Berger, with a 20 percent interest, were the general partners. Lowthian, with an 8 percent interest, was Freeland’s designee and was a limited partner. The HAB Trust and the MLB Trust, each with a 10 percent interest, were Berger’s designees and were limited partners. The final 20 percent was a limited partnership interest held under the name of Lake Murray Trust No. 1. The interests of the HAB Trust and the MLB Trust in SBIC were created by Sam Berger for his two sons, Harvey A. Berger and Marshall L. Berger. Sam Berger was the sole trustee of the HAB Trust and the MLB Trust. Neither the beneficiaries of these trusts, Harvey A. Berger and Marshall L. Berger, nor the trusts themselves held any interest in LMDC. The 20 percent interest in SBIC, held under the name of Lake Murray Trust No. 1, was established for the benefit of Jack Barenfeld and Samuel Glaser. At a later date, Barenfeld and Glaser gave 10 percent of their interest [579]*579in this trust to the other members of the Barenfeld-Glaser group.
“The preamble to the SBIC partnership agreement provided that the purpose of the partnership was to engage ‘in the business of buying, selling and developing land upon the real property commonly known as the Third Annexation of the Waring Property consisting of approximately four thousand five hundred (4500) acres of land in the City of San Diego, State of California.’
“The body of the agreement provided, inter alia:
“(a) That the term of the partnership would be 10 years.
“(b) That each partner, general or limited, was prohibited from selling, assigning, or transferring his interest in the partnership to any person other than another partner but that the agreement would subsequently be amended to resolve the question of the withdrawal or sale by the partners of a partnership interest.
“(c) That Freeland and Berger were to contribute $70,000 and $30,-000, respectively, to the partnership but further that they could withdraw said sums from the first monies available to the partnership.
“(d) The general business policy of said partnership in all questions relating to the management of its business should be determined by the mutual consent of the general partners.
“(e) Both general partners were to have full management rights and control.
“(f) Freeland was to receive an engineering fee of $25 per lot (in addition to any other engineering fees) from the first 2,000 lots ‘developed or sold by the partnership to third parties.’
“(g) That in the event the general partners could not agree, the problem would be submitted to arbitration.
“After the closing of the Kensington escrow on October 26, 1954, Berger was anxious to get LMDC activated. The engineering partnership and Freeland, as consulting engineer for LMDC, commenced applying for the necessary permits and talking with various city departments in order to get necessary city approval. Freeland’s engineering partnership was also employed by LMDC and performed various services for LMDC relating to on-site development and off-site improvements.
“On November 8, 1954, Freeland forwarded to the San Diego Planning Commission on behalf of LMDC a ‘Tentative Map Unit No. 1 [Lake Park Development Property].’ Unit No. 1 consisted of approximately 50 acres to be subdivided into approximately 260-280 lots. Submitted at the same time were ten copies of a master plan map for the entire 4,500 acres and ten copies of a master plan map of the 800 acres subject to LMDC’s 500 acre option. The master plan map of the 4,500 acres was an identical copy of that prepared for Kensington and Mrs. Waring on October 23, 1953, except that a portion of the lower right-hand section of that map had been cut out and a new section representing Unit No. 1 inserted. The map for the 4,500 acres indicated tentative locations for schools, parks, and supermarkets for the entire tract. It showed that, when fully developed, the 4,500 acres would yield an estimated 16,000 lots. As a result of discussions with representatives of SBIC and LMDC concerning approval of the subdivision plan, the representatives of the San Diego Planning Commission understood that Unit No. 1 was the first unit in a planned development that would ultimately total 4,500 acres.
“On November 4, 1954, LMDC purchased 3.19 acres by a partial exercise of its 500 acre option. LMDC then commenced construction of seven model homes on this acreage. It was necessary for LMDC to bring off-site improvements to the Lake Park Property’s southeast border in order to develop this [580]*580acreage. In late December 1954, the seven model homes were completed and opened to the public for inspection. The advertising campaign referred to a ‘City of Tomorrow’ to be developed on the 4,500 acres. Freeland, though he was aware of such advertising, disapproved of it and so informed Berger.
“On December 14, 1954, the City Council, on the recommendation of the Planning Commission, gave preliminary approval to the subdivision plan for Unit No. 1. Prior to December 23, 1956, LMDC, through its agent Freeland, submitted to the City Planning Department a tentative subdivision map for Units Nos. 2 and 3. Prior to January 17, 1955, LMDC similarly submitted a tentative subdivision map for Units Nos. 4 and 5. The size of the lots ranged between one-fourth of an acre and one-fifth of an acre. It was anticipated that 500 acres would produce between 2,200 and 2,400 lots. Units Nos. 1 through 5, if fully developed, would contain a total of approximately 3,000 lots and covered the 800-acre parcel on which LMDC had a 500-acre option.
“The City, in considering a tentative subdivision plan, was concerned, inter alia, with the location and adequacy of schools, parks, shopping centers, and off-site improvements, including adequate water supply. Though an immediate proposal might concern only a very small number of acres, the city considered the development in terms of overall plans for the surrounding area. The city would not permit wasteful duplication of facilities such as two small water reservoirs where one larger reservoir would be considerably more economical.
“Both Berger and Freeland represented SBIC in negotiations with the San Diego Unified School District pertaining to the granting of options for school sites within Units 1 through 5.
“On March 8, 1955, SBIC, through a letter signed by Berger, offered the School District school sites at a price equal to $2,000 per acre plus ‘the average off-site cost for the entire 4,500 acres of approximately $200 per acre,’ plus the cost of on-site improvements. Additional land for park and recreation purposes was offered on the same terms. This offer was not accepted by the School District. On September 23, 1955, SBIC, through a letter signed by Berger, offered the School District, for $1 each, options on school sites within Units 1 through 5 at $1,900 per acre for usable land. SBIC also agreed to give the School District renewable options on specific sites of land selected by the School District on the remaining areas to be developed by SBIC.
“On July 26, 1955, the San Diego City Council adopted the Final Subdivision Map of College Ranch Unit No. 1, which map was filed and recorded in the County Recorder’s Office on July 27, 1955. In July and August of 1955, LMDC partially exercised its 500-acre option and purchased an additional 146.93 acres from SBIC. LMDC then commenced development of Unit No. 1. LMDC also commenced the off-site improvement work for the 500-acre development. The principal aim in construction of these facilities was to provide adequate facilities for the 500 acres with minimum damage to the remaining 4,000 acres.
“In December 1954 or January 1955, a series of disputes arose between SBIC and LMDC. As a result of those disputes, LMDC and SBIC entered into an amended option agreement covering the 500 acres. The purchase price of the land was changed to $2,000 per acre for the first 250 acres and $1,800 per acre for all additional acreage. The ‘loan’ of $200,000 by LMDC to SBIC was la-belled a ‘credit,’ with LMDC entitled to recover any portion not used for off-site improvements, and no interest would be paid on the credit. LMDC’s prior oral agreement to pay Freeland $25 per lot for his personal service and counsel was reduced to writing. The agreement also eliminated the lot and off-site improvement cost guarantees by SBIC.
“LMDC engaged in the construction of 196 homes. By the summer of 1955, LMDC developed financial difficulties. [581]*581In an attempt to remedy the situation, Berger sought a change in the amended option agreement permitting LMDC to acquire what he thought would be more salable land. He proposed to develop land in the foothills of the mountain which would have a view of the lake. Employees of the engineering partnership prepared a feasibility study of the proposal, the conclusion of which was that the cost of off-site improvements would be prohibitive at that time.
“Ultimately, LMDC was unable to overcome its financial difficulties. It became insolvent on May 4,1956. LMDC assigned to Phoenix Insurance Company of Hartford all of its assets, except that it retained the right under the amended option agreement to purchase 200 acres of land out of the 500 acres originally covered by the option. Since LMDC had already exercised its option to the extent of approximately 150 acres, the portion of the option assigned to Phoenix only covered the right to purchase the remaining approximately 150 acres. Phoenix subsequently exercised this option.
“Berger was blamed for mismanagement of LMDC and was himself in financial difficulty. In August 1956, Carlos Tavares, representing a group of developers, including himself, approached Berger and offered to purchase his interest in SBIC. Tavares had been one of the developers approached by Freeland at the time of the annexation of the 4,500 acres.
“On August 10, 1956, the Tavares group agreed to buy out Berger’s interest and the respective 10 percent interests of the HAB Trust and the MLB Trust in SBIC for $950,000. Berger did not get Freeland’s consent for the sale of his interest in SBIC as required by the partnership agreement.
“Tavares then approached Freeland in an effort to purchase his interest in SBIC or, in the alternative, to have Freeland join the Tavares group in the development of the SBIC land. Freeland told Tavares he was not interested in joining in the development and decided to sell out. Lowthian decided to sell out her interest as well. Freeland received $730,000 for his interest, while Lowthian received $220,000 for hers. In the agreement of sale, they consented to the purchase from Berger.
“During the taxable periods portions of the acreage were utilized for grazing and the raising of barley. The partnership reported gross income from these sources in 1955 of $10,467.99 against total expenses of $27,609.92 (of which $15,319.51 represented real estate taxes) and in 1956 of $17,958.50 against total expenses of $45,032.69 (of which real estate taxes represented $15,619.01).
“Freeland did not wish directly to engage in subdividing and developing the land, either individually or through any other business form, because it would place him in competition with his clients and those of his engineering partnership and thus be financially injurious.
“At no time material hereto was either Freeland or Lowthian individually a dealer in real estate or individually holding real estate for sale to customers in the ordinary course of business.
“Although no improvements were made directly to the acreage not covered by the option to LMDC, the entire tract benefited from the off-site improvements made by LMDC. SBIC itself never directly advertised, promoted, or otherwise engaged in active efforts to solicit for its land, nor did it authorize anyone to undertake such activities directly on its behalf.
“Freeland, on behalf of SBIC, refused several unsolicited offers for the sale of portions of the acreage not covered by the LMDC option.
“SBIC never purchased any real property except the 4,500 acres here in issue. The only sales of real property ever made by SBIC, aside from the 300 acres acquired under the 500-acre option to LMDC, were (1) the sale of a fractional acre to the telephone company and (2) the sale of approximately four acres to the City of San Diego for a reservoir site as required by Kensington when SBIC [582]*582purchased the land. The only other land transactions of SBIC involved the grant to the San Diego School District of options to purchase sites within the area to be developed by LMDC for three elementary schools and one junior high school. SBIC also agreed to grant further similar options to the School District when development proceeded beyond the initial 500 acres. At the time of the annexation of the 4,500 acres, the city had required the then owners to agree to sell school sites within the property, and Kensington passed this obligation along to SBIC.” C.T. 116-33 (footnotes omitted).
On the basis of these facts the Tax Court further found that SBIC had held its land "primarily for sale to customers in the ordinary course of [its] trade or business.” This finding was made in spite of contentions by petitioners that the land had been held only for “investment” purposes. It was recognized by the court that none of the “normal criteria” relevant to questions such as the one involved here — for instance, frequency and continuity of sales, advertising, solicitation, extent of development and the like — could be relied upon in this instance to establish a purpose to sell to customers. The opinion went on, however, to state that:
“The normal criteria are not the exclusive benchmarks for decision in a case such as this, where there were good reasons for indirect as opposed to direct participation in development activities and where the period of gestation was prematurely cut short.” C.T. 146.
The court relied instead upon a theory of “indirect participation” in sales and development activities:
“We think it clear that SBIC did not itself intend to develop the 4,500 acres —such action would have been damaging to Freeland’s professional activities and those of his engineering partnership. But it is equally clear that the ultimate success of the venture, upon which SBIC had embarked, was dependent upon insuring the development of the acreage by someone. * * *
“ * * *
“ * * * We think that from the beginning, the partners in SBIC intended to sell off the property as quickly as the area could be made ready for development and that to this end they planned actively and continuously to participate in sequential development efforts. LMDC was the primer for the entire project. It was the first vehicle to be utilized and the partners of SBIC were deeply involved in the activities of that entity. Cf. Ackerman v. United States, 335 F.2d 521 (C.A. 5, 1964). Respondent suggests that LMDC was to be the sole vehicle of the partners, transferring its activity from one segment of the tract to the next as soon as the development of the previous segment was completed. We do not find it necessary to go that far. It is enough for us to say that the partners would seek constantly to construct vehicles for development, in which they would probably participate in one form or another, in order to be in a position to sell the acreage involved in slices of varying sizes as promptly as possible. Alternatively, they were ready to sell at any time that the partnership received acceptable offers. Cf. James E. Kesicki, 34 T.C. 675 (1960).” C.T. 139, 146-47.
Three points should be made initially in connection with our consideration of this general finding. First, if it was properly made, then the petitioners’ joint challenge to the Tax Court’s legal conclusions must be rejected. Assuming that the nature of SBIC’s business was in fact what the Tax Court found it to be — that is, that the partnership’s purpose was to have the land in question developed — then it would be immaterial that the partnership sought to utilize a legally separate entity, LMDC, in order to achieve its commercial ends. Cf. Ackerman v. United States, 335 F.2d 521 (5th Cir. 1964), and Tibbals v. United States, 362 F.2d 266, 271-272, 176 Ct.Cl. 196 (1966), in which development [583]*583activities of taxpayer-controlled close corporations, relating to land previously transferred to the corporations by the individual taxpayers involved, were held attributable to those taxpayers. (Indeed, in Ackerman the transfer was made indirectly — first to a trust and only later to the corporation.) In cases such as this, as in a great many situations, “the tax law deals in economic realities, not legal abstractions , * * Commissioner of Internal Revenue v. Southwest Exploration Co., 350 U.S. 308, 315, 76 S.Ct. 395, 399, 100 L.Ed. 347 (1956). Second, as with all findings of fact, we are compelled to uphold the trial court unless the finding at issue is “clearly erroneous.” E. g., Rollingwood Corp. v. Commissioner of Internal Revenue, 190 F.2d 263, 265 (9th Cir. 1951). And third, the evidence should be scrutinized in light of the fact that the burden of proof in the Tax Court was on petitioners. Tax Ct.R. 32; Factor v. Commissioner of Internal Revenue, 281 F.2d 100 (9th Cir. 1960), cert. denied, 364 U.S. 933, 81 S.Ct. 380, 5 L.Ed.2d 365 (1961). With these considerations in mind, we find ourselves unable to uphold petitioners’ contentions.
There was sufficient evidence, first of all, to support a finding that the original purpose of SBIC was that attributed to it by the Tax Court. That evidence includes Berger’s sweeping statements concerning imminent development of the entire tract; similar language in the letter to General Petroleum; the requirement in the contract with Kensington that SBIC deposit $200,000 in escrow for off-site improvements, which the Tax Court reasonably viewed as “mirror [ing] the vision of development which the parties, including SBIC, had in mind,” C.T. 139; the increasing amounts of the yearly payments required of SBIC under that contract, indicating (the Tax Court was entitled to find) that development may have been intended to be the source of repayment funds; the “interlocking” nature of SBIC and LMDC membership; the provisions in the original SBIC-LMDC contract which in effect passed some of the latter’s risks and opportunities for profit in development activities on to SBIC; the ./discrepancy between the $1037 per acre paid SBIC and the $2000 per acre it immediately received from LMDC (which, though certainly explicable in part on other grounds, was not unreasonably taken to represent to, some extent “a method of siphoning off to the common participants in SBIC and LMDC a portion of the latter’s profit hopefully as long-term capital gains,” C.T. 144); and the fact that the off-site improvements to be provided out of the $200,000 deposited by LMDC could be viewed as “not only * * * for the benefit of the initial 500-acre tract but also [for the indirect benefit of] the remaining acreage,” C.T. 144-45.
True, the inferences drawn from some of these facts were not required inferences, and there was evidence in petitioners’ favor with regard to many issues. The Tax Court might possibly have properly found in petitioners’ favor. It may be that no one piece of evidence in itself would support that court’s finding. In combination, however, the factors referred to above provide a substantial basis for the finding of SBIC’s original purpose regarding the land.
Petitioners contend, however, that the relevant intent for purposes of the issue here raised is that held immediately prior to sale. See Tibbals v. United States, 362 F.2d 266, 176 Ct.Cl. 196 (1966). And they vigorously urge that even if the purpose found by the Tax Court may be attributed to SBIC at the outset of its operations, that purpose changed in light of later events. They point, first, to the disputes that occurred between SBIC and LMDC; second, to the failure of LMDC; and third, to the sale of Berger’s interest in SBIC to Tavares.2
[584]*584Petitioners’ contentions admittedly are not wholly without support. There perhaps would have been some basis for a finding that SBIC’s original purpose changed after the occurrence of these events. The nature of the questions is such, however, as to make it difficult for us to rule in accord with petitioners’ contentions. As a general matter, first of all, courts in situations such as that involved here should be wary of finding that, because of a lack of business success, the purpose for which property is held by a partnership changes from one of sale to customers to one of investment. The clear purpose of section 751 is to prevent members of partnerships from reducing their tax liability by converting potential ordinary income —attributable to inventory or unrealized receivables — into capital gain by selling their partnership interests. Roth v. Commissioner of Internal Revenue, 321 F.2d 607, 611 (9th Cir. 1963), and sources cited therein. This purpose would in many instances be frustrated if partners’ decisions to sell their interests necessarily sufficed to establish the sort of “change of purpose” which petitioners contend occurred here. Moreover, “intent” is, particularly when applied to an impersonal entity such as a partnership, to a great extent a fiction — or, periaps more accurately, a legal construct. It is conceivable that, particularly after "the failure of LMDC and the entry into SBIC of Tavares (who wanted to develop -the land directly), SBIC’s “purpose” regarding the land was to some degree not clearly defined. Freeland did not want to involve himself directly in .development activities, yet the instrument which the partnership had chosen to use as a means of engaging in “indirect” development — LMDC—no longer existed, and one of the original partners — -Berger—was in financial difficulty. Courts must, however, often decide questions which are thus speculative in nature. And, considering (1) the supported finding that SBIC had at first intended to develop the land indirectly, and (2) the fact that the burden of proof was on petitioners, this court cannot conclude that it was “clearly erroneous” for the Tax Court to find that the partnership’s purpose had not changed. We cannot hold as a matter of law that that purpose must have changed.
Two further issues are raised by petitioner Lowthian. She contends, first of all, that the “intent” of the partners engaged in indirect development activities through LMDC ought not to be attributed to her. For her, she argues, the SBIC venture was merely an “investment.” This latter assertion may, of course, be true. But it is the intent of the partnership, and not of any individual partner, that is in issue according to the Internal Revenue Code. Under the written partnership agreement, to which petitioner Lowthian was a party signatory, the business policy of SBIC was to be determined by the general partners, Berger and Freeland; Lowthian, as a limited partner, thus agreed that she should have no hand in determining that purpose. The case of United States v. Rosebrook, 318 F.2d 316 (9th Cir. 1963), aff’g 191 F.Supp. 356 (N.D.Cal.1960), cited by petitioner, is inapposite. There, the taxpayer sold a one per cent interest in land which had originally been purchased by her father in the name of a trust having her as beneficiary. Although the interest had been committed to a joint venture, it was directly owned by the taxpayer, who was a tenant in common; under section 1221 of the Code, therefore — and especially since she had succeeded involuntarily to ownership of the land and participation in the joint venture — the court looked to her purpose rather than to that of other participants in the venture, and ruled that she had held the land as an investment. Here, both essential elements of the factual situation in Rosebrook — direct ownership and involuntary participation in the joint scheme — are absent. Petitioner’s tax liability is clearly governed by the purpose of the partnership.
Finally, petitioner Lowthian contends that the Commissioner abused his [585]*585discretion in reopening with respect to her 1956 liability. The Commisioner asserts that he acted under Rev.Proc. 59-25, 1959-2 Cum.Bull. 938 (since superseded by Rev.Proc. 63-9, 1963-1 Cum.Bull. 488, and Rev.Proc. 64-40, 1964-2 Cum.Bull. 971), which states,
“It is the administrative practice of the Internal Revenue Service not to reopen cases previously closed by the District Director unless there has been substantial error, both in amount and in relation to the total tax liability, or there is evidence of fraud, malfeasance, collusion, concealment or the misrepresentation of a material fact.”
Pointing out that there is no evidence of “fraud, malfeasance, collusion, concealment or the misrepresentation of a material fact,” and that the Commissioner must therefore have acted under the “substantial error” clause of the provision, petitioner contends that the latter phrase refers not to the substantiality of the amount involved, but to the “obviousness” of the mistake claimed to have been made. Since the factual question here involved is a relatively close one, she concludes that reopening was not authorized. We reject petitioner’s proposed construction of the clause, however, not only because it might impose unwise and impractical restrictions upon the Commissioner’s power to reopen, but also because the very language of the provision in question clearly defines “substan-tiality” in terms of “amount” and “relation to the total tax liability.” There is no contention that petitioner Lowthi-an’s deficiency is not substantial in either respect, and, indeed, any such contention would be bound to fail.3
Affirmed.