BELL, Chief Justice.
Appellees recovered judgment against appellant for $176,945.76 with interest thereon, and attorneys’ fees in the amount of $12,-500.
Appellees by their first amended original petition, on which they went to trial, alleged that by contract dated September 1, 1950, and accepted September 25, 1950, entered into on behalf of all appellees by Meredith, Clegg and Hunt and McCarthy Chemical Company, predecessors of appellant, appellees agreed to sell and McCarthy Chemical Company agreed to purchase, all natural gas and gas distillate or condensate that should be produced from wells in the West Hamshire area belonging to appel-lees. The part of the contract material to the disposition of this case reads as follows :
“You (McCarthy) shall run said production to your plant full stream, in which event quantities of gas and distillate run from the above mentioned Unit shall be determined on the basis of reasonable and proper tests for such proper purpose made by Southern Petroleum Laboratories, or some reputable testing organization selected by you and approved by Meredith Clegg and Hunt. All tests shall be run at seven hundred fifty (750) pounds separator pressure and at seventy (70°) degrees Fahrenheit temperature, and full stream volumes shall be corrected for deviation from Boyles Law and to sixty (60°F) degrees Fahrenheit temperature. A test run shall be made at the time such well is put on stream and first runs are made into the pipe line. Thereafter, such tests shall be run quarterly. The volume of gas and quantities of distillate, as determined by each test, shall be the basis of measurement for all gas and distillate delivered into the pipe line for the next succeeding three (3) months period. However, should the results of two successive tests differ, the volumes of gas and distillate delivered during the period intervening between such tests shall be considered to be the average volume of the two tests — that is to say that the average volume shall be one-half (½) of the total of the two volumes from the two tests — and on determination of such average volumes an adjustment of accounts shall be made between the parties in the next payment to be made by you under the terms of this agreement.”
In October, 1950 appellees allege they commenced delivery from the E. C. Han-kamer No. 1 and that a test called for by the contract to determine the distillate content of the gas delivered was made by Southern Petroleum Laboratories, the testing agency named in the contract. As a new well was connected with appellant’s plant, it would be tested in the same manner by Southern Petroleum Laboratories. There were three other wells from which gas was taken: Weisse No. 1, Weisse No. 2 and Federal Royalty. Thereafter, a like test was to be made quarterly. It was alleged that on each well an orifice meter was
installed by appellant to measure the gas produced by the well, all of which gas went to appellant. The meter was installed and maintained by appellant and appellant placed, changed and read the charts in the orifice meter as provided in the contract. Appellees allege that each month representatives of appellant, as provided in said contract, would read the charts from the orifice meters and calculate the amount of gas delivered for the preceding month from each well, and, taking the distillate content of the sample of the gas taken, as determined by the test made by Southern Petroleum Laboratories, appellant determined the amount of distillate that had been delivered to appellant the preceding month. Appellant would then send a statement to Meredith, Clegg and Hunt, the operator for all appellees, showing the amount of gas and gas distillate sold to appellant the previous month from each well, and this statement would be accompanied by a check of appellant. For the first quarter after a well had been placed in production, the distillate content for a given well would be determined by the initial test made on the well by Southern Petroleum Laboratories. At the end of each quarter after deliveries from a well had commenced a new test would be made by Southern Petroleum Laboratories and the average of the two tests would be made and the average would be applied to determine the distillate delivered in the intervening time between the two tests and there would be adjustments upward or downward in the accounts, as the case might be. Too, at the end of each quarter appellant would send its office meter charts to Meredith, Clegg and Hunt who checked the charts to correct any errors that might have been made in reading the charts and in calculating the amount of gas and distillate delivered. This method of operation continued through September, 1952.
Appellees then allege that during the month of September, 1952, they sold to appellant gas and distillate from the four wells above named, in the aggregate amount, based on contract price, of $110,-960.11, but that during the month of October, this being the month for which accounting was to be made for sales during September, appellant withheld the sum of $87,232.59. It is then alleged that in October appellees sold distillate and gas at contract price' in the aggregate amount of $116,766.45, and that appellant withheld the sum of $87,232.60 in the accounting month of November. Then it is alleged that during November gas and distillate in the amount of $16,734.49 was sold but that appellant withheld $2,480.57 in December. In each instance the monthly deliveries from each well are shown as in an itemized account and the pleading is sworn to as an itemized account for materials sold and delivered.
Appellees then allege the contract to be clear and unambiguous and that as a matter of law the parties agreed to be bound by the results of the split stream tests run by Southern Petroleum Laboratories to determine the distillate content of the gas, and the parties agreed, as a matter of law, to accept and be bound by payments made on a basis of these split stream tests made by Southern Petroleum Laboratories.
In the alternative, appellees allege that if the contract is not clear and unambiguous, the parties by their conduct as above alleged for a 20 months period construed the contract to mean they were each bound by the split stream tests made by Southern Petroleum Laboratories. It is alleged that if the contract requires any explanation the only explanation necessary to make the terms and provisions plain and unambiguous is a construction of the language with respect to the tests to be made. Then appellees say the contract calling for tests to be made at a temperature of 70° Fahrenheit and a pressure of 750 pounds necessarily called for the split stream test. Too, they allege the parties knew that the only testing equipment Southern Petroleum Laboratories had was equipment to make split stream tests and that the only equipment then in use on wells supplying gas to
appellant was split stream test equipment. The conclusion is pled that on a basis of the contract provisions the construction and intei-pretation of the parties and the fact that it was known at the time the parties entered into the contract that the split stream test was universally used by Southern Petroleum Laboratories in testing other wells in the area for appellant, as a matter of law and fact, the split stream test was to be used and the parties were to be conclusively bound by such test as made by Southern Petroleum Laboratories in calculating the distillate content of the gas and the amount of distillate to be paid for.
Appellees then prayed for attorneys’ fees alleging the suit to be one on a sworn account for materials furnished, to wit: natural gas and distillate. Claim for attorneys’ fees was made under Article 2226, Vernon’s Ann.Texas St.
Appellant filed a sworn denial of the correctness of the account. In addition it specially pled that in June, 1952, it discovered that it had paid approximately $180,000 for distillate which it had not received. The basis of the assertion was that due to the installation by appellees of sampling loops of incorrect design and installation more distillate was shown by the tests run by Southern Petroleum Laboratories to have been delivered than was in fact delivered. The deficiencies in the sampling loops were alleged to have been unknown to appellant.
Additionally, it was alleged that shortly after June, 1952, when it was determined that erroneous results were being obtained by the split stream method of testing, due to the use of improper testing loops, it was mutually agreed that adjustments would be made if it were determined by subsequent tests that adjustments were due. It is then alleged that subsequent tests conclusively established that adjustments were due in the amount of $180,000. This agreement allegedly resulted from a letter of Meredith, Clegg and Hunt to appellant dated July 19, 1952, and a letter from appellant in response thereto dated July 22, 1952, and a so-called letter contract of September 16, 1952.
It should also be noticed that appellant filed its plea in abatement contending that the Texas Company was a necessary party to the suit as it owned an interest in the Federal Royalty well. This plea in abatement was overruled.
Trial was to a jury, but at the conclusion of the testimony, the Court withdrew the case from the jury and rendered judgment for appellees in the amount above stated.
Appellees discharged their burden of proof by introducing in evidence the monthly statements prepared by appellant showing the gas and gas distillate delivered during the months of September, October and November, 1952, months concerning which there is no dispute with regard to the amount of gas and gas distillate delivered, after proving them up. It was from the amounts due under these statements that appellant had deducted the amounts above stated on the theory that because of errors in the tests made by Southern Petroleum Laboratories to determine the distillate content of the gas previously delivered there had been an overpayment, that is, more distillate had been paid for than had been delivered. While liability for attorneys’ fees was contested, it was agreed that if there was any liability the sum of $12,500 was a reasonable attorney’s fee.
Evidence introduced showed that the gas and distillate from the four wells passed from the wells full stream directly to the processing plant of appellant, where it and gas and distillate from other owners’ wells were all mixed and processed. An orifice meter was installed and maintained by appellant’s representatives and the charts which recorded the necessary data for determining the cubic footage of gas that flowed into appellant’s plant were placed in the meter and removed from the meter by appellant’s representatives. The cubic footage of gas delivered was determined each month by representatives of appellant
from the data shown on these charts. Representatives of appellant would then take the result of the tests made by Southern Petroleum Laboratories to determine the distillate in the gas and by multiplying the number of barrels of distillate which by these tests had been found to be in a million cubic feet of gas by the number of million cubic feet of gas delivered they would obtain the amount of distillate delivered in a given month. Representatives of appellant then prepared statements showing the amount of gas and distillate delivered from each well, and also showing the amount due appellees at the contract price, and sent the statements monthly to Meredith, Clegg and Hunt, the operators of the wells, together with appellant’s check covering the payments thus shown to be due. These statements together with payment for the products were made the month following the delivery of the products. The contract provides for payment on or before the 20th day of the succeeding calendar month.
Periodically these charts are sent to Meredith, Clegg and Plunt and their representatives check the charts for any error in reading them and in calculations made from them.
As each well was put in production Mr. Frier of Southern Petroleum Laboratories made a test of the gas to determine the distillate content. The result of this test was used for the next three months to determine the total gas distillate delivered from each well. Then another test would be made on each well and if there was a difference shown between the tests as to the distillate content of the gas an average of the two tests would be taken and applied for the deliveries made during the preceding quarter and adjustments made at the next billing. Some such adjustments were made from time to time.
This method of operation continued from October, 1950, when the Hankamer No. 1 -well was placed in production, until June, 1952.
The test employed by Mr. Frier of Southern Petroleum Laboratories was what is known as the “split stream” test. In making this test a sample of the full well stream was taken and it was run through a separator at a pressure of 750 pounds and at a temperature of 70 degrees, as called for by the contract, and the distillate in the sample separated from the gas. The quantity of gas employed in the sample was measured and the distillate separated from it was measured. From this by simple multiplication the number of barrels of distillate in a million cubic feet of gas could be determined. In order to obtain a representative sample of the full stream a small pipe called a “sampling loop” was inserted in the pipe line near the well and through this loop a sampling tip was inserted into the full stream. The tip in the stream in effect cuts a plug of the stream and the sample thus obtained is run through a miniature separator.
What has been said to this point, we think, is without dispute in the evidence, and we believe all we have stated is the effect of evidence admitted. We note this because most of the record before us is testimony offered but excluded on objection. We will seek to differentiate between that admitted and that excluded.
Appellant called Mr. Frier of Southern Petroleum Laboratories to the stand, he being the person who made all of the tests involved in this case. Mr. Frier testified there is the same degree of accuracy in a split stream test and a full stream test. There is a recognized tolerance between the two tests of 2 per cent. When the Hanka-mer No. 1 well came in in July, 1950, he made a test of the well. Fie made another test in October, 1950. The October test was the initial test on the Hankamer well provided for in the contract. (All of the tests made were split stream tests unless we indicate otherwise). The July 1950 test was 'made under very adverse conditions and the size of the sampling loops is not shown. The witness thought it was a loop belonging to McCarthy. He consid
ered his October, 1950 test to be the most accurate (He had made a third test in November but it is not material here). When he made the test in October he could see part of the sampling loop. In the July, 1950 test he used the wrong shrink factor. He corrected it back to what he thought it ought to be. This test was run under very adverse conditions. He stated there could have been a sizeable error in the July test that affected the liquid content of the gas. He could not put a limit on his error. He could have used the wrong tip. He could have missed it, but doesn’t know how far because he was testing under various adverse conditions. On the second (October) test, he had plenty of time. Present at this test, among others, was Mr. Miller, General Manager of McCarthy Chemical Company. In the second test the same loop was used, so far as he knew, as in the first test. The second (October, 1950) was more accurate than the first (July, 1950). In the November test he thought the loop had been cut off because he had trouble getting the sample tips through the valve. The loop was buried. The second test was made under normal conditions. His opinion as to the October test being the most accurate is based on the conditions under which it was made. After the initial test he made quarterly tests, using the split stream method. The results of the tests were sent to both parties, meaning the purchaser and representatives of sellers. The reports of each test on each well were introduced by appellant for the limited purpose of showing inaccuracies. (This in order to lay predicate for later showing the difference between the gas paid for and that in fact delivered as shown by the proffered testimony of Ashbury Parks.)
In June of 1952 Mr. John Ashford, of Texas Gas Corporation, asked him to make a test. He made the regular quarter split stream test and some full stream tests. The two wete not run simultaneously. The purpose was to compare results. The results between the full stream test and the split stream test varied more than the 2 per cent, tolerance recognized. The full stream was not run at 750 pounds pressure and temperature of 70°. This was unnecessary since the split stream was run under the same conditions for purpose of comparison. They were seeking to determine if they were getting a representative sample in the split stream tests. After these tests he saw the sample loop that had been used in the Federal Royalty well and he would call it a “short” loop. He did not see the loops used in the other three wells. In July, 1952, after the sampling loops theretofore used had been dug up and replaced! by longer loops, as he had been informed, Mr. Frier made subsequent split stream, and full stream tests. In these tests the-split stream and full stream were run simultaneously and under identical conditions, though not at contract conditions,, of 750 pounds pressure and 70 degrees-temperature. They were run at line conditions. The results of the tests were the-same except in the Federal Royalty well. The results on it didn’t check for two tests,, but in subsequent tests did.
Carl Cane testified for appellant that during the time here material he was employed by Meredith, Clegg and Hunt. He installed the sampling loops on all of the-wells. No specifications were furnished him on two of the loops. He constructed; these two of the loops from available material. The loops were not of uniform, size. The loops were buried. In July,. 1952, he removed the loops, or at least he-helped a contractor move them. They were replaced by the installation of other-loops. The witness was not asked anything about the dimensions of the loops.
Mr. John Ashford, an oil and gas consultant, who was employed by appellant and' its predecessor at times material to this case, also testified. Much of his testimony is devoted to a description of the split stream-test and the full stream test. He testified the only way to test the accuracy of the split stream test is to run it under identical conditions and simultaneously with the full; stream test. The accuracy of the split
stream test is dependent on the use of a representative sample of the stream. A full stream test can be run at 750 pounds pressure and 70 degrees temperature. The witness was then asked about tests made in June, 1952. Objection was made by appel-lees to any testimony concerning special tests because the contract called for tests to be made by Mr. Frier of Southern Petroleum Laboratories. Further, the evidence showed Mr. Frier made the split stream tests for twenty months and appellants accepted his results and testimony of tests made by others without consulting appellees would not be admissible as to whether the contract was performed. Further, there was no evidence that the tests were not in accordance with the contract and the parties are bound by the tests made by Mr. Frier. The objection was sustained and the following testimony was given on the bill of exception.
The substance of the witness’ testimony which was excluded was that after studying the various tests made by Mr. Frier there seemed to be wide fluctuations that should not exist and the quantity of products produced by appellant’s plant seemed deficient. It was then determined to make special tests. Mr. Frier, in his presence, and in the presence of some of appellees, ran full stream tests and split stream tests. The original sample loops were used. The tests were run simultaneously and there was a variance between the results, the split stream test showing approximately 4 barrels of distillate more per million cubic feet of gas than was shown by the full stream test. A rough calculation showed that, based on an estimated monthly delivery of gas, there was an overpayment of $225 per day, or $6,500 per month. The tests were not run at 750 pounds pressure and 70 degree temperature, as called for by the contract. After these tests the testing loops were removed and longer ones installed by appellees.
In July, 1952, after the sample loops had been changed, additional tests were run by Mr. Frier at which representatives of appellant and some appellees were present. Split stream tests and full stream tests were run simultaneously and the result of these tests was that the distillate content of the gas as shown by the split stream tests was in substantial agreement. After this Mr. Asbury Parks was employed to study the previous tests made by Mr. Frier and determine if they were in error and if so to what extent. Mr. Parks’ report did not itself show the extent of error of any dollars and cents, but showed what the tests should have shown during the prior test periods. After receiving this report from Mr. Parks, calculations were made showing the extent of over-payments. There were then offered the documents showing adjustments alleged to be due under the report of Mr. Parks, these alleged adjustments having been sent Meredith, Clegg and Hunt. These were excluded.
After the July tests, meetings were had between representatives of appellant and appellees. At one such meeting Mr. Frier stated the errors in the tests made in June were due to the short sampling loops of which he was not aware and after these had been removed and replaced with longer loops the split stream tests and the full stream tests were in substantial agreement. Mr. Asbury Parks witnessed these tests.
Mr. Asbury Parks was called as a witness by appellant. He testified he designed and built the first equipment to be used in a split stream test. He described this type of test. He stated the difficulty in using this test was in getting a representative sample of the well stream. If a representative sample is obtained in a split stream test, this test will give the same result as a full stream test. Use of a full stream test is standard in the industry for evaluating the accuracy of the full stream sample used in a split stream test. Split stream tests are used primarily as a matter of economy.
From this point testimony given by Mr. Parks was excluded. He testified on the bill of exception that the obtaining of an accurate sample is basic. He was employed by appellant in June, 1952 to make an analysis of the split stream tests made by Mr.
Frier from July, 19S0 through September, 1952. He had available the tests made by Mr. Frier in June, July and September, 1952. The report on the June, 1952 test made by Mr. Frier indicated the sample loops were not delivering correct samples. The same sampling loops had been used in previous tests so it was his opinion there was error in the sample obtained. The witness had witnessed no tests prior to July, 1952. The tests made by Mr. Frier in July and September, 1952 confirmed that the sample loops used prior to July were not yielding representative samples of the well stream.
In the July tests both split stream and full stream were used. The tests showed sub-’ stantially the same results as to distillate content. It was his opinion that samples from longer loops produce more reliable samples. The difference in the June and July, 1952 split stream tests was in his opinion due to the difference in the sampling loops. The only thing that had been done to the testing system was to change the loops. He was unable to fix the degree of error in the samples taken from October, 1950 through June 1952, but he was only able to say the samples were unreliable and would not be used in computing the distillate content of the gas. He was able to make “a reasonably accurate determination” of the true liquid content of the well stream since production. In making his determination of the distillate content of the gas delivered from October, 1950 through June 1952 he analyzed every report Mr. Frier made. The first test considered was that made in July, 1950. He considered this after considering Mr. Frier’s comments regarding sample loops. He checked Frier’s calculations and he found no reason to believe there was any substantial error in this test (July, 1950). In this test Mr. Frier knew the type of sampling loop he was using. In making his determination of the liquid content of the gas delivered from October, 1950 through June, 1952 Mr. Parks discarded all tests made by Mr. Frier during such period as unreliable because the sampling loops used during this time were found to be in error in June, 1952. The only test he had no reason to doubt was the July, 1950 test on the Hankamer No. 1. He checked the calculations on this report and made a very slight adjustment in liquid content. This was due to his use of a different table. The tests of July and September, 1952 wire of unquestioned accuracy since the split stream test of this period was lun simultaneously with the full stream test. He then averaged the July and September results and took this as the true liquid content of the well stream as of that time. He then projected a line back from this figure toward the figure indicating the liquid content as shown by the test of July, 1950. For lack of any other factual information of any nature he connected these points with a straight line. This was as to the Hankamer No. 1. For the Weisse Nos. 1 and 2 and Federal Royalty wells, for lack of any reliable information, he took the July, September average and projected a line parallel to the line for the Hankamer well to the point where they were put in production. From this he calculated the distillate content of the gas during the intervening period. Then appellant offered tests made on the wells by Mr. Frier covering- the period from the cancellation of the contract here involved in November, 1952 through 1955. These reports showed the split stream test was used and checked by the full stream test. These tests confirmed his determination of the distillate content of the gas for the period from October, 1950 through June, 1952.
The effect of the summary of the record above given is that on the testimony admitted by the court appellees were entitled to recover the amounts withheld by appellant in October, November and December, 1952, because there is no question concerning the amount of gas and distillate delivered during the months which the statements for these months represented, and there is no evidence admitted which would indicate any overpayments for previous months because of deficiencies in delivery, the most being there was a possi
bility of error, but none admitted showing the extent of error, if any.
The excluded evidence above summarized was offered to establish the fact that there were mistakes made by Mr. Frier in the tests he made from October, 1950 through June, 1952; that the mistakes resulted from the use of improper sampling loops which resulted in inaccurate samples of the well stream being obtained; that this caused it to appear that there were greater amounts of distillate delivered than were in fact delivered; that appellant had paid for distillate not delivered, and the extent of such overpayment.
Appellant has assigned 100 points of error. Other than the points relating to admissibility of evidence, the allowance of attorneys’ fees and the exclusion of evidence alleged to establish a new contract with regard to settlement of the dispute alleged to have been made in 1952 after the dispute arose, appellant’s basic contentions may be stated thus:
The purchase contract called for the measurement of the distillate in the gas by the use of "reasonable and proper” tests to be made by the Southern Petroleum Laboratories; that the contract does not make the tests run by the testing agency conclusive and binding on the parties and it is therefore permissible to prove that the tests as run in fact produced erroneous results which resulted in appellant paying for more distillate than it received and it is entitled to restitution of this overpayment. Appellant says that the undisputed evidence in the record (that admitted and that offered but excluded, but which should have been admitted) showed that errors in measurement during the period from October, 1950, through June, 1952, had been made in the tests made for the purpose of determining quantities of distillate, which errors were due to the use of sampling loops of improper design and construction; and, that the loops were installed and buried by appellees or persons other than appellant, without knowledge of appellant or Southern Petroleum Laboratories. The proof, says appellant, also showed the errors resulted in overpayment in the amount withheld by it. Further, it says these errors in testing prevented the tests from being “reasonable and proper” tests called for by the contract. Also appellant says these errors being unknown to the parties at the time payment was made establish a mutual mistake.
Appellant says if these matters were not established as a matter of law at least the evidence raised fact issues as to their existence which should have been submitted to a jury.
Appellant does not contend, nor did it plead, fraud or bad faith on the part of appellees or Mr. Frier of Southern Petroleum Laboratories, the testing agency in performance of the tests. It merely contends it was entitled to show, and did show, that a mistake had been made in testing which resulted in its paying for distillate it did not receive and it is entitled to restitution.
Without again summarizing the excluded evidence, it is sufficient to say that appellant contends the above summarized evidence going to show erroneous measurements, the cause of the errors and the extent of them, all of which was excluded, should have been admitted.
The appellees respond that all excluded evidence was properly excluded because the purchase contract called for testing by use of the split stream method and does not allow use of the full stream method to check the accuracy of the split stream method; that the use of the split stream test was employed with the knowledge of the parties for 20 months and payments made on a basis of the tests made and the parties themselves have thus mutually interpreted the contract as calling for testing by the split stream test. Too, they contend that the contract called for testing by Southern Petroleum Laboratories, and this made the testing agency their mutual agent and it was intended that the tests made by this agency should be final and
conclusive on the parties. Too, they say that the parties thus interpreted the contract by accepting the tests as run and paying on a basis of such tests for 20 months of the contract. They say that the contract having made the tests of Southern Petroleum Laboratories conclusive on the parties, these tests may not be impeached in the absence of fraud or bad faith.
We have reached the conclusion that the Trial Court correctly withdrew the case from the jury and rendered judgment for appellees for the amount that was withheld by appellant from the amount due for gas and distillate delivered by appellees during the months of September, October and November, 1952.
No one disputes that the split stream method of testing is a reasonable and proper test. The real question on this phase of the case is whether the tests as actually run are binding on the parties regardless of errors (assuming there were errors) so long as there is no showing of fraud or bad faith.
We must determine the intention of the parties. We think unquestionably the use of the term “reasonable and proper” referred to the type or character of the test to be employed, but not the manner in which it was conducted. Admittedly the split stream test is a reasonable and proper test used in the gas industry for determining the distillate content of natural gas containing distillate in quantity.
We hold, however, that by the terms of the contract the parties intended that the tests as run by Southern Petroleum Laboratories should be final and conclusive on the parties and there being no contention or showing of fraud or bad faith in manner of performance, they may not be attacked as being merely erroneous.
It is elementary that in construing a contract the intention of the parties is to be determined. In the absence of ambiguity, and there is none here, the intention is to be determined from the language used by the parties, the subject-matter of the contract, the situation of the parties and the surrounding circumstances. 10-A Tex. Jur., Contracts, § 177, p. 357; Ryan v. Kent, Tex.Com.App., 36 S.W.2d 1007; Gibson v. Turner, 156 Tex. 289, 294 S.W.2d 781; Cudlipp v. C. R. Cummings Export Co., Tex.Civ.App., 149 S.W. 444.
The subject-matter of the contract here involved was natural gas saturated with small droplets of substantial quantities of distillate, the distillate having value separate and apart from the gas because of its properties. Because of its being suspended in a large stream under great pressure it was no simple matter to determine how much distillate the gas contained. Measurement required the services of an expert.
Now let us look at the terms of the contract.
First, it provided that the production (meaning the gas and the distillate it contained) should be run to appellant’s plant full stream. Here it would be processed. The distillate would lose its identity .as such because products would be made or manufactured from it. Secondly, it provided for reasonable and proper tests to be performed by “Southern Petroleum Laboratories or some reputable testing organization selected by you (appellant) and approved by Meredith, Clegg and Hunt (the operators for ap-pellees).” The parties selected Southern Petroleum Laboratories and thus made it their common agent. Not only this, but if any other testing agent were employed it must be reputable and it must be approved by both parties. Whoever was to make the tests was to be their common agent. Thirdly, when a given well was put on stream1 and the first runs were made into the pipe line an initial test was to be made to determine the distillate content of the stream and this test “shall be the basis of measurement” for the next three months. All that would then remain' to be done would be to apply the contract price to determine how much appellant owed. The
term “basis of measurement” could but mean that this test would be used to determine the total amount of distillate contained in the gas delivered for which payment was to be made. After determining the total amount of distillate in all the gas •delivered there only remained the matter of applying the contract price to determine the amount due for a given month. Fourthly, payment was due on or before the 20th ■of each month following delivery. The significance of this is that payment was to follow after the product had been proc■essed and its identity thus destroyed. Fifthly, it provided that each well should be tested quarterly and if there was a difference in the last two tests the average of the two would be taken and the “average volume of the two tests, that is to say that the average volume shall be one-half of the total of the two volumes from the two tests and on determination of such average volumes an adjustment of accounts shall be made between the parties in the next payment to be made under the terms of this agreement.” This average was to be applied to the deliveries made for the previous three months. Thus the parties gave recognition to the possibility of error in a given test and made provision for adjusting accounts by providing for averaging two tests and applying it to deliveries already made.
We do not mean to say that the selection of a common agent of itself suffices to make his determination of a given matter final and conclusive, but this together with other circumstances may be sufficient to show that the parties intended a common agent’s determination to be final.
Appellant denies that the contract either expressly or by necessary implication makes the test by the testing agent final and conclusive on the parties. However, it seems to concede that the parties may so contract and if they do they are bound in the absence of fraud or bad faith. We agree that the rule is that if parties to a contract select a common agent to determine the quantity of a product and expressly, or by necessary implication make his determination final, the determination by the agent, in the absence of fraud or bad faith in making the determination, is conclusive on them. Cudlipp v. C. R. Cummings Export Co., supra; Blum Milling Co. v. Moore-Seaver Grain Co., Tex.Com.App., 277 S.W. 78; Booth-Kelly Lumber Co. v. Williams, 95 Or. 476, 188 P. 213; Hayday v. Hammermill Paper Company, 176 Minn. 315, 223 N.W. 614, 63 A.L.R. 210.
The case of Cudlipp v. C. R. Cummings Export Co., supra, presents such a well-reasoned opinion and a fact situation so closely analogous to this case that we shall especially notice it. It definitely supports our conclusion that the' contract here involved, by necessary implication, though not expressly, makes the tests of Southern Petroleum Laboratories conclusive on the parties, and our conclusion as to the rule of law that where such is the case the tests may be impeached only by showing fraud or bad faith. In that case Cudlipp sued to recover the value of logs he claimed were sold to and received by the defendant for which he had not paid. The contract was oral. It was established without dispute. The logs were to be put in the river by plaintiff and left to float down the river to the mill of the defendant. While on the skidway at the defendant’s mill the logs were to be scaled and classified by defendant. They were then cut into lumber at the defendant’s mill and shipped out. Settlement was to be made on a basis of the scaling and classification thus made. The jury found against the plaintiff, but the opinion of the court is such that we feel had the jury not so found, judgment would have been rendered for the defendant as a matter of law. The trial court had charged the jury that the parties were bound under the contract by the scaling and classification made by defendant unless there was such gross mistake as to show fraud or bad faith. By such charge in effect the court told the jury the contract made the defendant’s determination conclusive though the
contract did not so expressly provide. The plaintiff contended since the contract did not expressly make the determination conclusive, that the scaling and classification could be impeached by proof of mistake alone. This Court, as then constituted, held that the terms of the agreement were such that it was to be inferred from the language used, the situation of the parties and the subject-matter of the contract, that the parties intended to make defendant’s determination conclusive. In reaching such conclusion the court particularly noticed that the contract provided for settlement on a basis of such and that the logs immediately following scaling and classification were sawed into lumber and shipped out of the mill.
We have read each case cited by appellant and deem them inapplicable in the light of the situations materially and substantially differing from the situation before us. For illustration, in each instance the article, the quantity or quality of which was to be determined, was, so far as we recall, still available for further checking. To particularly distinguish each in this opinion would serve no useful purpose but would merely lengthen a necessarily long opinion.
The trial court, therefore, correctly excluded all testimony offered for the purpose of showing the fact of mistakes in testing by Southern Petroleum Laboratories and designed to show the extent of alleged overpayments. Without such testimony the uncontradicted testimony showed that during September, October and November, 1952 appellees had delivered products to appellant for which they had not paid.
From what has been said, it is apparent that we are of the view that no evidence was admissible which sought to show that the tests made by Mr. Frier were in fact erroneous,-or that sought to show the extent of such error and the allegedly resulting overpayment, because the contract made his tests final and conclusive on the parties and there is no contention of fraud or bad faith on the part of appellees or Mr. Frier. There are, however, additional reasons why we hold the Trial Court correctly excluded the proffered testimony. We need not state all of them, nor itemize the evidence excluded.
First, evidence was offered with a view to establishing the fact that a mistake was made by Mr. Frier because the sample loops were too short. There was offered the result of the June, 1952 tests where the split stream test and the full stream test were run under the same line (not contract) conditions and there was a wide variation between the distillate content as shown by the split stream test and the full stream test. This was offered, not to show the extent of error, but the fact of error, it having been testified elsewhere (though the evidence was excluded) that the two tests should produce substantially the same result, there being a recognized tolerance of 2 per cent. It was really offered to show a representative sample was not being gotten, a representative sample being basic to the accuracy of the split stream test. This of itself shows only that something was wrong. However, other excluded testimony showed the sampling loops in use during the time material to this suit were then removed. After they were replaced, allegedly with longer loops, the split stream test and the full stream test were run simultaneously, at line conditions (not contract conditions), and produced substantially the same results, at least within the recognized tolerance. From this Asbury Parks and Mr. Ashford concluded the difficulty lay in the loops being used. A letter from Mr. Frier pointed out that in the earlier tests short loops had been used and they were replaced with longer loops though no view was expressed as to the effect of such. The loops were changed, but we have been unable to find any admissible testimony showing in fact that longer loops were installed as every witness seemed to say he had been informed this was true. We feel, however, that the showing of a variance in the tests of June, 1952, when the old loops were in use, and a coincidence in results
from the split stream test and the full stream test after the loops had been changed, in the absence of some other explanation, would reasonably lead to the inference that the loops were the cause of an error, and thus a fact issue as to the fact and cause of error would be raised. Appellees’ argument that no one, even Mr. Frier, had ever previously suggested the dimensions of the loops was material and no one ever testified as to the proper dimensions, goes to the weight to be given the above testimony. The evidence does show that witnesses offered by appellant were not even asked about the proper dimensions, when such witnesses were in a position to know. Appellees argue that because they were in a position to give the facts and did not creates a situation where the above testimony concerning the June and July tests, has no efficacy as a matter of law. We feel this failure to produce testimony would not as a matter of law destroy the above mentioned inference, but would be a fact to consider in determining the ultimate issue of the cause of the variance between the split stream and full stream tests. However, we feel the evidence inadmissible for the reason first above stated; but apart from such reason, we think it would have been admissible and sufficient to raise a fact issue as to the fact and cause of error, certainly with the added conclusion of the experts that this was, in their opinion, the cause of error.
However, no harm could have resulted to appellant because it still labored under the burden of establishing the extent of error; that is, how much distillate was in fact delivered over the previous 22 months. They had this burden because they had previously paid for whatever had been delivered and this suit was to recover for distillate undeniably delivered in September, October and November, 1952, and not for that allegedly delivered prior to July, 1952.
We are of the view that the evidence offered to show the extent of error and thus to show the amount alleged to have been overpaid was properly excluded on grounds in addition to that previously discussed.
Mr. Asbury Parks was the witness primarily relied upon to establish what distillate had in fact been delivered from October through June, 1950. He testified that he took an average of the distillate content of the Hankamer well as shown by the July and September, 1952 tests. The testimony shows that in these tests the split stream and the full stream tests were run simultaneously at line conditions, that is, at approximately 900 pounds pressure and 86 degrees temperature. These were not the contract conditions. The contract expressly called for the distillate content of the gas to be determined by tests run at 750 pounds pressure and 70 degrees temperature. The distillate content for these months, as we read the record, was actually determined by the distillate being separated by the separator at line conditions and then, theoretically, the distillate content was adjusted to what Mr. Parks thought the distillate content would have been had the test been run at 750 pounds pressure and 70 degrees temperature, the specifications called for by the contract. How he did this we are not informed. He did not, as we understand it, merely use the full stream to see if a representative sample was being obtained by use of the new loops and then have a split stream test run at contract conditions. This July-September, 1952 average was Mr. Parks’ beginning point. The vice we see in it is that it was employing a standard not prescribed by the contract, but in fact at variance with it. On a graph, for the Hankamer well, Mr. Parks marked a point showing the distillate content as of that date. Having concluded that all tests made by Mr. Frier beginning in October, 1950 and ending in June, 1952 were unreliable, he discarded them. He selected the test made by Mr. Frier in July, 1950, when the Hankamer well was first put in production. The record shows that Mr. Frier testified he considered this test not reliable because it was made under very adverse conditions. Too, we note the test
was made at 720 pounds pressure and 70 degrees temperature. This was not the contract pressure and though there was a difference of only 30 pounds, we do not know what difference it would make and the record is silent on this. Then Mr. Parks on the graph marked the distillate content as shown on this (July, 1950) test. It should be noted that Mr. Parks did not witness this test. Later Mr. Frier discovered he had used the wrong “shrink factor” and corrected the report to show a greater distillate content. This was done after Mr. Parks had made his determination, and he never made adjustment for this and there was, so far as we can tell, no adjustment made for this by Mr. Parks.
Then Mr. Parks testified that for want of any reliable information (he having discarded all tests made by Mr. Frier for the intervening period) he connected the two points with a straight line. From this he determined what he thought was the distillate that was in fact delivered from October, 1950 through June, 1952. He then passed this information to appellant whose representatives calculated how much in dollars and cents had been overpaid under appellant’s theory. As to the other three wells the average of the July and September tests was taken and a point put on the graph. Since Mr. Parks had no previous test he considered reliable, he arbitrarily drew a line parallel to the line drawn for the Hankamer well from the point showing the July-September average back to the date each well was put in production.
We think the effect of the above procedure was to substitute a method for determining the distillate content of the gas for that prescribed in the contract. This may not be done.
Appellant offered tests made for succeeding years through 1955 in an effort to establish the accuracy of his. determination for the period from October, 1950 to June, 1952. As we read the record, all such tests were made by using the split stream and full stream tests at line conditions with an alleged adjustment to contract conditions. This evidence was thus correctly excluded.
Appellant contends the Texas Company was an indispensable party to the suit as it owned an interest in the Federal Royalty well.
The Texas Company was not a party to the contract sued upon. After the execution of the contract, Texas Company executed a pooling agreement whereby it acquired an interest in the Federal Royalty well which was completed in January, 1952. In March, 1951, Texas Company, and the plaintiffs named in the suit, entered into an operating agreement with Meredith, Clegg and Hunt whereby the latter became the operator of the Unit. Texas Company also executed a division order in favor of Meredith, Clegg and Hunt authorizing it to produce and sell, and collect the money for gas and distillate. In addition to this, W. A. Mudgett, an attorney for Texas Company, testified he was authorized to appear and testify and that the attorneys representing appellees were authorized to sue for Texas Company in the name of Meredith, Clegg and Hunt, and to collect any sums due Texas Company.
Under the record we hold Texas Company (Texaco) was not an indispensable party.
Meredith, Clegg and Hunt were authorized to sell gas and distillate for Texas Company, to collect for that sold and remit to Texas Company. The contract sued upon was made between appellant’s predecessor and the appellees. Texas Company was not a party. However, Meredith, Clegg and Hunt sold the gas and distillate of Texas Company to appellant by virtue of the contract. Under authority of the division order Texas Company was an undisclosed principal, unknown when the contract was made. Too, the agent contracted in his own name. Further, the usage of the trade authorizes an agent to act as owner of the gas and distillate in the light of the division order. Also, the agent has
an interest in the subject-matter of the contract. The firm of Meredith, Clegg and Hunt was the agent. Under any of these four circumstances the agent may sue in his own name. M. W. Fruit Co. v. Bierbauer, Tex.Civ.App., 216 S.W.2d 831, writ ref., n. r. e.; 2 Tex.Jur., p. 577.
There was no agreement consummated by the letters above referred to. There was merely an agreement to try to work out the differences between the parties. The letters were not admissible as evidence of the interpretation of the contract by the parties. Such would be admissible only if the contract were ambiguous, which it is not, which appellant concedes. We do not, however, think it constitutes any interpretation by the parties.
Appellant lastly contends the court erred in allowing attorneys’ fees, because the cause of action arose before the amendment of Article 2226, Vernon’s Ann. Texas St., allowing recovery on a sworn account, and there can be no retroactive effect given the statute. It also contends that the gas and distillate sold by appellees to it are not “materials furnished.”
In its first contention appellant is correct. Government Personnel Mutual Life Ins. Co. v. Wear, 151 Tex. 454, 251 S.W.2d 525, and Moutos v. San Saba County Peanut Growers Ass’n, Tex.Civ.App., 268 S.W.2d 761, no writ history.
The statute, however, at times material to this suit, authorized recovery of attorneys’ fees where there was recovery of judgment for materials furnished.
We are not unaware that the courts have held the statute to be penal in nature and that they therefore construe it strictly. Moutos v. San Saba County Peanut Growers Ass’n, supra, and Davenport v. Harry Payne Motors, Inc., Tex.Civ.App., 256 S.W. 2d 245, no writ history.
No cases have been cited by either party that are determinative of the question. We have reached the conclusion that gas and gas distillate as furnished to appellant come within the meaning of materials furnished, as used in the statute. Webster defines “material” thus: “The substance or substances, or the parts, goods, stock or the like, of which anything is composed, or may be made; as raw material.” The products here were not sold as is feed to be consumed, or gasoline to be consumed, or as is an implement employed to construct but which does not become a part of the thing made from it. The products were sold to be processed at appellant’s plant. Products were to be made from at least some of the constituent elements of the gas and distillate. The gas and distillate here sold were materials furnished just as surely as lumber, bricks, cement and nails that go into the construction of a building.
The judgment of the Trial Court is affirmed.
WERLEIN, J., not sitting.