Poole Engineering & MacHine Co. v. Swindell

Offutt, J., delivered the opinion of the Court. Walter B. Swindell, Charles J. Swindell, and Walter B. Swindell, Jr., co-partners, trading as Swindell Brothers, have for many years been engaged in the manufacture and sale of bottles at their factory at Bayard and Russell Streets in the City of Baltimore. The process of manufacturing the bottles which they sell consists of reducing certain materials to a mass of molten glass and by means of machinery transforming that mass into bottles of divers sizes and shapes. In the course of it the molten glass is discharged from a feeder, in lumps or gobs approximately uniform in size, into moulds in a bottle-making machine through which by the application of condensed air it is forced into the desired shapes. Tlie machinery employed in the operation is extremely complicated, and its efficiency depends largely upon its capacity to maintain the flowing glass at fixed temperatures for definite periods through the several stages of manufacture. The bottle-making machine used in the final stage of manufacture is heavy, durable and designed to stand the wear and tear of continuous use under working conditions over considerable periods of time. The moulds are more fragile, have a comparatively short life, under usual operating conditions must be frequently replaced, and are sufficiently expensive to constitute a material element in the economical operation of the business. They naturally vary as the size and shape of the bottle to be made varies, and, as the number of bottles of a given shape made from a single set of moulds increases, *574 the expense of the moulds allocated to each hottle decreases. The machines ordinarily used by the appellees prior to the occurrences referred to in the declaration in this case were designed to operate with twelve moulds. The expense of making so many moulds to produce a comparatively small number of bottles of a given size or design necessarily made the cost of each bottle proportionately higher than it would have been had a greater number been produced. For that reason the use of a machine which would operate with four moulds would be more economical in producing a small number of bottles of a given size and shape than the use of one which would not operate effectively with less than six or more. There was, therefore, in the bottle manufacturing business, a need for a machine which would operate effectively with a small number of moulds, and Clarence Reuben Nixon, an employee of the appellees, undertook to design and construct such a machine, and at the same time to test on it a device which he had invented for producing a more perfect bottle than those ordinarily made on standard machines. At that time, what might be called the standard method of manufacture, in making each bottle, utilized two moulds, the receiving mould into which the gob of molten glass was first discharged by. the feeder, and the final or “blow mould,” into which it was transferred after its treatment in the receiving mould had been completed. Nixon’s device introduced an intermediate stage, by having the forming bottle transferred from the receiving mould to a forming mould before it was finally completed in the blow mould. In that method the1 first mould is called a “preliminary blank or pattern,” the second the “forming blank or pattern” and the third the “mould.” He finally constructed a machine having four moulds and utilizing the device which has been described, but which was so made that, in order to preserve the ratio of time and temperature to the volume of the flowing glass essential.to the moulding process, it was necessary to alternately drop one gob from the feeder into the receiving mould and one into a reeeptable called a “basement,” so that one-half the material discharged by the feeder was wasted. That waste could be *575 avoided if the bottle-making machines were of snch a size that two of them could be operated in connection with a single feeder, and when so operated such machines would effect- a material economy in filling “small orders.” There was evidence in the case tending to show that the Nixon machine when completed adequately demonstrated its own practical utility, as well as the value of the Nixon device, and the appellees were so impressed by its possibilities that they formed the design of having it strengthened, perfected, manufactured, and sold to the trade. While they manufactured their own moulds and did much of their own repair work, they were not equipped for the construction of such machines, so that, while that which Nixon constructed was adequate for experimental purposes, it was not apparently strong enough to stand the strain of prolonged operation under -ordinary working conditions, and appellees felt it necessary to have the machine, which they had in mind, manufactured by some company regularly engaged in the manufacture of machinery. Subsequently negotiations began between them and the Poole Engineering & Machine Company, the defendant below, and the appellant here, for the construction of such a machine. Whether they were initiated by them or by tbe appellant does not clearly appear and is not material, but it does appear that they began prior to March 5th, 1928, and that on May 11th, 1928, one of the appellees addressed the following letter to the appellant: “I have just returned from a convention at Atlantic City and find your letter of May 8 before me, hence the delay in answering. “I am very glad you are progressing so satisfactorily with the drawings and general lay-outs for the new machine. In accordance with your request, I enclose one of our order blanks on which we give you authority to proceed with this work for us, and this will become a matter of record with your order department.” Following that letter, the appellant, hereinafter referred to as “Poole,” undertook the construction of a bottle-making machine for the appellees at the cost of construction plus *576 twenty per cent, thereof, and in July, 1929, it delivered what purported to he such a machine at the appellees’ factory. The machine was subjected to different tests, the last,of which was made on December 27th, 1929, and on March 21st following that test Poole demanded of the appellees payment of a balance claimed by it to be due for work and materials furnished in the- construction of it, amounting to $1,431.64. In reply to that letter the appellees stated that the machine was worthless, that they rejected it, and they not only refused to make any more payments on account of it, but demanded that Poole refund to them payments they had made to it, aggregating $23,881.64, on account of the construction of the machine, and upon the failure of Poole to comply with that demand, on September 8th, 1930, brought this action in the Baltimore City Court to- recover such payments, which in an amended bill of particulars filed in the case were shown to aggregate $23,701.51, made as follows: “June 10, 1928, $241.41; July 13, 1928, $194.30; Aug. 11, 1928, $461.63; 'Sept. 13, 1928, $427.05; Oct. 12, 1928, $175.19; Nov. 16, 1928, $2,863.12; Dec. 28, 1928, $1,-330.39; Jan. 24, 1929, $2,999.57; Eeb. 21, 1929, $7,504.56; May 1, 1929, $4,354.09; May 3, 1929, $1,317.81; June 17, 1929, $816.09; July 9, 1929, $699.61; Aug. 12, 1929, $316.69.” The case was. in due course tried before the court and a jury, and the verdict and judgment being for the plaintiffs for that amount, the defendant appealed. At the close of the whole case the plaintiffs offered one prayer and the defendant nine, of which the first, second and third were granted, and the others refused. The defendant also- specially excepted to the plaintiffs’ prayer, but its exception was overruled and the prayer granted. Those rulings are the subject of the single exception presented by the record. The case was tried upon antithetical theories. The contention of the plaintiffs was that the defendant had contracted to- manufacture for and sell to them a machine for a known and definite use, that as part of the contract it warranted the fitness of the machine for that use, that relying *577 upon the warranty they made payments from time to- time on account of the purchase price of the machine, that the defendant, although given every reasonable opportunity to do so, failed to perform its contract, but tendered in pretended compliance with it a machine which was wholly unsuitable for such use aud worthless, and that because of that breach of its contract and warranty they were entitled to recover the payments they had made. The defendant’s contention was: (1) That the evidence in the case was legally insufficient t-O1 permit any inference other than that the only contract between the parties was one for labor and material, and not one of sale; (2) that, even if there was a contract of sale, the evidence was legally insufficient to support the hypothesis that the defendant had warranted that the machine when completed would have a known and described utility; (3) 'that, if there was such a warranty, the evidence was legally insufficient to allow a finding that the completed machine failed to comply with its terms; (4) that, even though the original contract contained a warranty of fitness, nevertheless, if the plaintiffs subsequently assumed control of the design of the machine, such conduct discharged the warranty, if the completed machine conformed to a design adopted, created, or approved by them. The first "and the most important consideration which occurs in reviewing the court’s rulings in reference to the plaintiffs’ prayer, the defendant’s demurrer prayers, and the special exceptions, is whether there was in the case evidence legally sufficient to support the hypotheses: (1) That the defendant contracted with the plaintiffs to “manufacture and sell” a machine; (2) that the defendant promised that such machine would have a given utility, to wit, that of making “merchantable bottles in merchantable quantities”, and (3) that the completed machine was incapable of making “merchantable bottles in merchantable quantities.” A second question is whether, if there was evidence legally sufficient to entitle the plaintiff to recover at all, the plaintiffs’ prayer fairly stated the conditions precedent to1 such a recovery under the facts of the case. *578 While the first question demands a searching and critical examination of the evidence, it would be impossible in an opinion of any reasonable length to do more than refer to such' parts of it as are essential to the conclusion reached as a result of such an examination. A great part of it dealt with the details of the construction and operation of bottle-making machinery, which, while perhaps necessary to' a clear understanding of the case in the trial court, has no necessary connection with the questions presented in this court. Much of it is expressed in technical language or trade vernacular in connection with exhibits before the trial .court, and is, apart from those exhibits, which are not available to this court, unintelligible. The real issues in the case were: (1) Did Poole contract with appellees to manufacture for and sell to them a bottle-making machine, which (2) would make merchantable bottles in merchantable quantities, (3) did it perform that contract, (4) if not, was its failure due to any want of skill or care on its part or to some fault in the specifications which it followed at the direction of the plaintiffs ? In referring to the evidence relevant to those issues, it will be stated partly in narrative form, and in connection with the plaintiffs’ prayer, the special exceptions thereto', and the defendant’s demurrer prayers, and for that purpose, and for that purpose only,, so much of it as tends to support (a) the plaintiffs’ right to recover, or (b) the supposed fact that the design or plan of the machine conformed to directions given by the plaintiffs, will be assumed to be true. There had been some discussion between representatives of the appellees and Poole concerning the manufacture of a bottle-making machine by the latter at least as early as March 5th, 1928. By April of that year, Mixon’s work on the experimental machine had so far progressed that appellées were convinced of the practical utility both of the machine which he had constructed and of the device which he had invented. But to completely utilize the economical possibilities of that machine, it was necessary that it be of such a size that it could be economically used in combination with another machine of like size in connection with a single *579 feeder. ' The reason for that was this: The Nixon machine operated four moulds, but in order to maintain the temperature of the flowing glass at a point, essential to the manufacture of merchantable bottles, it was necessary that it flow in sufficient volume to feed eight moulds, and that at the same time there be a sufficient interval of time, between the discharge of the gob of molten glass into the receiving mould and its transfer to the forming blank and then to the blow mould, to permit the heat in the interior of the mass to penetrate to the exterior and soften a film or shell caused by its contact with the cooler iron of the receiving mould. To preserve that balance of time and temperature, it was necessary, first, that the moulds be operated at a speed low enough to permit that reheating, and, second, that the volume of flow through the feeder be great enough to maintain the temperature required in the molten glass. The problem presented by these opposed but essential requirements was solved in this way: The volume of the flow needed to maintain the proper temperature was sufficient to feed eight moulds, but there were only four moulds on the Nixon machine. Nixon, therefore, so adjusted the feeding apparatus on his machine that it would cut off and deliver one gob to' the receiving mould, and cut off and deliver another to a receiving “basement.” In that -way he preserved the ratio of time between the stages of the moulding process, as well as the necessary volume of flow, but he wasted one-half of the raw material. To avoid that waste he proposed to operate two machines in connection with the single feeder, so that it would alternately drop one gob into a mould on each machine, instead of dropping a gob alternately into a mould and into a receptacle for waste. But that method of operation was not practicable unless the over-all size of the machines was substantially the same as that of the Nixon machine. The appellees therefore wanted a machine made which would be approximately of the dimensions of the Nixon machine, would embody its essential feature, would function in the same manner, but which would be sufficiently strong and durable to stand the *580 strain of prolonged, operation under ordinary working conditions. Following antecedent. correspondence and interviews, in April, 1928, representatives of Poole visited the Swindell plant to inspect the Nixon machine, with a view of submit ting a bid for the manufacture of such machines as the appellees required. On that occasion appellees informed Poole’s representatives that there was a question as to “what kind of person we should employ to build the machine, whether a bottle machine specialist, and we pointed out the various difficulties-connected with bottling machinery, the fact of the heat involved, the expansion and contraction, and we tried to impress what we considered was the difficulties of a bottle-manufacturing machine.” At that time appellees had in use other standard machines known as “Lynch” machines, and, after examining those machines, Clarence M. Davison, chief engineer for Poole, assured appellees that the question (as to their ability to do the work) was “perfectly ridiculous” and “practically said” that the Lynch machines “were a piece of junk, they could see right off the bat how they could change it and give us a greater and better production,” that it would be “a simple thing to reproduce” the Nixon machine, and Mr. Dudley Shoemaker, vice-president of Poole, who was present at the time, said it would take about three or four months to build the machine, and that it w'ould cost from $3,000 to $5,000, and “he took out a book and pencil and began to figure; I said, ‘How can you do it like that?’ and he said, ‘We have a method of computing according to weight, and we estimate that weight so and so much per pound or ton, and it should cost, oh, around three to five thousand dollars, and after the first "one is built, why, it might be as low as $1,500.’ ” Poole was “very anxious to get the job,” “were insistent that they were capable,” and, in answer to appellees’ expressions of doubt, assured them that it was nothing more than “a machine problem,” that there was “no difference between glass bottle machine manufacture • or their form of machine manufacture.” On the same occasion Poole was *581 informed of appellees’ plan to use two of the machines together in connection with one feeder. In respect to that, Walter B. Swindell, a member of appellees’ firm, testified: “Was anything ever said, to your recollection, and if so, to whom, and by whom, with respect to putting two machines under one feeder ? A. Unquestionably. Q. When was it? A. That was right at the beginning. Q. To whom was that said? A. That should have been heard by both Mr. Shoerqaker and Mr. Davison. Q. By both Mr. Shoemaker and Mr. Davison? A. Should have been. Q. By whom was anything said with regard to the two machines under one feeder? A. It was said by all of us, I imagine; I know I said it. * * * Q. Did you tell them about what you testified here yesterday about wanting to use this machine for small orders? A. Yes. Q. Could you tell them without explaining you intended tO' have two machines at the same outlet ? A. Ho. Q. Have you any doubt that you told them that? A. Xone. Xixon, referring to the same conference and speaking of the Xixon machine, said: “Q. What did he say he could do about making the same kind of machine in a strengthened and workable fashion A. He could give us a machine that would stand up satisfactorily, he could duplicate this machine and put it into condition whereby it could stand the strain. Q. Did he say he could make a machine that would produce commercial bottles? A. Yes, sir; he did.” And later the same witness said: “Q. Mr. Dixon, when did you first hear mentioned the two-machine idea in this matter, I mean in connection now with the Poole Engineering Company, or any employee of the Poole Engineering Company? A. The first day we all met in the factory, Mr. Davison, Mr. Shoemaker, Mr. Swindell and Mr. Harry Brawner. Mr. Dunnock, Lou Eickert and myself were there, we explained what we wanted was two single-unit machines. The machine we had there was to be duplicated and kept down to a small machine, so it could be used under a feeder, that was in the beginning. John L. Dunnock, an employee of appellees, referring to the same occasion, testified that Davison said that Poole could *582 easily reproduce the Nixon machine and “make one that would work.” Following those assurances, on April 30, 1928, Poole sent the appellees a letter in which it said: “I think that our engineers now thoroughly understand the requirements in the matter of the bottle-making machine that we have been discussing. “In accordance with our plan this morning, you will send the machine to us so that we can have it available for the purpose of checking when we make the full set of drawings, which will be necessary, of course, before we can proceed with the strengthened model. In the matter of making drawings, our Engineering Department will simply charge you straight draftsmen’s time for the work. Of course, it would be advisable to have, your men here with our people as the designs are developed. “On a job of this character, which is to some extent experimental and development work, it is impossible to estimate the exact cost, and' on the first machine we will charge you our factory cost plus twenty per cent., rendering bills monthly so that you can follow the expenditures as we go along. “Our boys are very enthusiastic over the machine and I feel quite sure that we can develop this proposition to your entire satisfaction and I am hopeful that it will lead to big future business.” That letter was followed by another dated May 8th, 1928, in which Poole said: “We plan to make a general lay-out proposition first in the drawing room and then from these lay-outs, which will be approved by your engineers, we will make detailed working drawings so as to proceed with manufacture in the shop. We will be glad to have an order from you to proceed so that we can file it in our Order Department.” In reply to that letter 'appellees on May 11th, 1928, gave the following order: *583 “Purchase Order “Swindell Brothers “Bayard & Bussell Sts. “Baltimore, Md. “Mr. Dudley Shoemaker “Poole Engineering & Machine Co. “Baltimore, Md. “Enter our order for the following, subject to the conditions mentioned below. “You are hereby authorized to make up the detailed working drawings for the new Mixon machine, and to proceed with its manufacture when these drawings are mutually approved by your engineers and our machine department.” On May 12th, Poole, in a letter of that date acknowledging the order, said, “Your Mr. Mixon was here on Thursday and our Engineering Department is now ready to proceed with the lay-outs in accordance with his ideas,” and on May 14th followed it with another in which it further acknowledged the order “to make up detailed working drawings for the new Mixon machine and also to proceed with its manufacture when these drawings axe mutually approved by our engineers and your machine department.” Mo complete set of drawings ever was furnished appellees, and on June 4th, 1928, they, in a letter of that date, complained “that no greater progress had been made on their machine.” On June 5th Poole answered, excusing the delay on the ground that Davison, their engineer, -was absent, and on September 22nd, Davison wrote to Shoemaker: “There is no problem about this glass bottle machine that we cannot work out properly, and to the final satisfaction of the customer. We must, however, be allowed reasonable time for the proper consideration of each important detail item before the final adoption of any scheme. * * * We have four men at work making up detail drawings and if everything goes smoothly, we will be completed within about three weeks.” *584 Up to that time, while there were available detail drawings of parts, there were none of the machine as it would appear after the parts had been assembled, and, while appellees were ready to co-operate with Poole, Poole had sent no one to further inspect the machines at the Swindell plant, and its engineer, Davison, said he did not need to “see bottles to construct a machine”; that he would “make a machine that would knock their eye out.” From September until the following January, Nixon and other employees of the appellees visited the Poole factory at irregular but not infrequent intervals, to observe the progress in the construction of the new machines, but could form no idea of what its appearance or dimensions would be when completed, or how closely it would reproduce the Nixon machine. In January, however, Nixon and Dunnoek were shown the base of the new machine and were astounded at its size, and so informed W. B. Swindell, Jr. He was very much disturbed by the information and promptly went to look at the base himself, and when he saw it at once realized that it would be impossible to get two machines of the size indicated by the base under one feeder, and so informed Davison. But Davison assured him that the Poole machine would double the output of the Nixon machine, and so produce the same result as two of the Nixon machines. Swindell then emphasized the fact that, if it would not do that, the Poole machine would be worthless to them, but Davison “insisted on it,” and showed him, “how, through taking advantage of what he figured to be the idle moments of the Nixon machine, he could double or more than double” its output. Even then Swindell was worried and hesitated “whether we would allow them to continue,” but Davison “said they (his doubts) were nothing; he could as well make a bottling machine as the people who were working on them all their lives; no question about the ability of the mould to make SO' many bottles. That seemed to worry us, because the number of bottles he was undertaking to deliver had never been done before, and has not yet been done.” *585 Mixon in his testimony gave this explanation of Davison’s confidence: “Undoubtedly, Mr. Davison didn’t understand that these idle periods he saw on our machine or the Lynch machine were due to the fact that they are the most essential periods in creating a bottle, that that period is known as reheating; after that glass has had contact with the iron, it is necessary that the heat go from the exterior to the interior, so that the veneer can be melted away until that all becomes at one universal temperature, before you dare insert air into it, for if you do insert air into it prior to the time it becomes evenly heated, naturally, the air pressure will go to the weakest point and it will break through the shell and your bottle will be thick and thin.” The dimensions of the Mixon machine were two feet eight inches in width by four feet in length, while the Poole machine was over five feet wide by seven feet two inches in length over all. After the assurances by Davison, Poole was permitted to proceed, and on January 15th, 1929, it addressed to the appellees a letter in which it explained that the delay was due to the time needed for the “preliminary lay-out and designing and making up the drawings from which to complete the machine in the shop”; that it had been necessary to complete over “three hundred working drawings, not including the preliminary lay-outs and free hand drawings, which it was necessary to get up with the help of you and Mr. Mixon before we could actually design the machine”; but that it would “start assembling” it “the latter part of next week” and that it would “be ready for you about the twentieth of February”; and concluded with the statement that “starting the end of next week we will want your people here almost constantly. * * *” After that, work on the machine proceeded, but it was not ready for delivery on February 20th, and the bills for that month and for March were not paid when presented, and, following a demand by Poole for payment, there was a somewhat acrimonious discussion, partly oral and partly written, concerning the cost of the machine and the delay in delivery. *586 The differences were finally harmonized, however, the work resumed, and at that time appellant again gave assurances that the machine “would he satisfactory,” “would be everything that was claimed.” It was finally delivered on July 16th, after Davison had told Nixon that when delivered it would be one hundred per cent, perfect. That, however, was not the case, and when, after tests conducted by Poole extending over several months, the machine failed to manufacture a single merchantable bottle, and after Poole, although given reasonable opportunity to do so, was unable to so improve its structural design as to make it possible for it to function adequately, the appellees rejected it. The Nixon forming blank which was used on the Poole machine functioned properly, not only on the Nixon machine, but also, after the Poole machine had been rejected, on Lynch machines, and in the opinion of Nixon the vital defect in the design of the Poole machine was that it ignored the relation between the volume of the flow of glass through the feeder and the time needed for moulding the molten glass. Although, as has been stated, there was evidence that Poole was to duplicate the Nixon machine, Nixon on cross-examination admitted that at the beginning he had suggested a rearrangement of the moulds, and one other change, which was adopted. The defendant offered evidence which, in addition to contradicting that of the plaintiffs in its most important and essential particulars, tended to prove these facts: When the construction of the machine was discussed by the parties to this appeal prior to the formation of any contract, nothing was said about its size, and when at that time appellant’s engineer was shown the Nixon machine, he was told that the location of the moulds on the Nixon should be changed and that they should be differently arranged on the proposed machine; that the size of the- Poole machine was determined partly by that specification, partly by appellees’ deinand that the new machine be so- constructed as to make larger bottles than could be made- on the Nixon machine, and partly by the incorporation in the new machine of a take *587 out device suggested by the appellees for removing completed bottles. When the Poole machine was about three-fourths completed, Exxon suggested a change in the apparatus designed to effect the final, blowing operation, which required a complete rearrangement of the timing of the several operations, and the “scrapping” of a number of parts wdiich had been made. In fixing the time ratios for the operations of the new machine, appeilant’s engineers were unable to' get any definite information from the appellees or their employees, and were compelled to rely upon their own experiments and calculations. But as work on it progressed, they from time to time showed drawings of it to Eixon, and also exhibited to him wooden models of parts, and he frequently suggested changes which appellant made. When he was shown the Eixon machine, Davison paid no attention to the fact that its operation resulted in a "waste of one-half the raw material, nor did he consider whether that waste was essential to the successful operation of the machine, and as to that he said: “I did not know that there is a minimum at which yon can make Bottles. I did not inquire about it. Eo one ever told me. I do not know what it is xiow. I don’t know anything about how long or how short. I don’t know how much time you must leave that bottle in the mould. I made the machine without any hypothesis as to that.” But, as will appear from the following extracts from his testimony, he did know that the purpose of the machine was to make bottles: “They wanted a machine that would make a bottle ? A.' I presume they did. Q. It would not be of any use if it did not make a bottle? A. Eo. sir. Q. You knew that was their object to get a machine to make bottles ? A. Yes, sir. Q. Commercial bottles? A. To make commercial machines to sell to the public. Q. In order to sell it to the public, they had to have one that could be used? A. Yes, six’. * * * Q. You knew, of course, the machine would not he of any use to Swindell or anybody else unless it made bottles ? A. I would expect it would not. Q. So it could not give satisfaction unless it made bottles ? A. I imagine that would be their idea, but *588 we were not interested in the glass-making part. Q. You would not expect them to he satisfied unless you made a machine that would make bottles? (Witness) They would not be satisfied with the machine unless it made bottles. (Mr. Markell) You would not expect that?. A. No, sir.” He later summarized appellant’s position as follows: “We had the Nixon machine as a guide and Nixon as an engineer to guide us. For the engineering part of this job we depended on Nixon; for the glass-making end of it we did. So far as our company is concerned, there was no professional engineering consideration given whatever to- the performance of this machine, not glass-making part. We gave no professional engineering consideration whatever to the making of glass by tljis machine, or as to whether the machine would make a bottle or not.” The same conception of appellant’s undertaking also appears in the following extracts from the testimony of Dudley Shoemaker, vice-president of the appellant company. “Did you- hear from any one during the progress of making this machine that, as Mr. Wilson testified here the other day, there was a time minimum of between six and seven seconds that the glass had to remain in the mould. Did you ever hear of any such limitation as that? A. No. * * * Q. In other words, a problem of that sort is one that must be worked out in the science of glass making ? A. That was their' problem, the Swindell Company’s and not ours. Our problem was to make a machine that would carry the moulds, turn the moulds, and run smoothly and not shake, and stand up as a machine. The question of time was out of our province. When Mr. Brawner and I made this agreement, it was a cooperative proposition, and their engineers were to take care of that, and ours of making the machinery. * * * Q. You never found out how long a bottle had to stay in the mould, did you? A. No, I don’t think that was'ever fully established by us. We had nothing to do with that part of the game. Q. How much do you think you learned about bottle-making, if you didn’t learn how long a bottle had to stay in the mould? A. We learned not very much. Q. And you *589 would have to know how to make bottles before you could make the machinery to make them? A. No, the' machinery is approved by the fellow who makes the bottles, and we make the machine.” In discussing the question as to whether those facts are legally sufficient to support a conclusion that appellant warranted the fitness of the machine which it undertook to construct for the use which it knew, when the contract was made, the appellees intended to make of it, much of their argument in this court was directed to determining whether the contract was one of manufacture and sale or merely one for labor and material. Assuming the truth of all evidence tending to support the appellees’ case and all inferences legitimately deducible therefrom (Griffith v. Benzinger, 144 Md. 599, 125 A. 512; Kelly v. Huber Baking Co., 145 Md. 333, 125 A. 782; Singer Const. Co. v. Goldsborough, 147 Md. 639, 128 A. 754), the contract in this case was obviously more than a mere contract for material and labor, under which the appellant was employed for no other purpose than to make and assemble the necessary parts of a machine in accordance with definite specifications. At a time when it was fully cognizant of the use to which the machine was to be put, Poole, in its offer made in its letter of April 30th, 1928, said: “I think that our engineers now thoroughly understand the requirements in the matter of the bottle-making machine. * * * On a job of this character which is to some extent experimental and development work, it is impossible to estimate the exact cost, and on the first we will charge you our factory cost plus twenty per cent. * * * I feel quite sure that we can develop this proposition to your entire satisfaction. * * *” And on 'September 22nd, 1928, Davison wrote to Shoemaker: “There is no problem about this glass-bottle machine that we cannot work out properly, and to the final satisfaction of the customer. * * * All general schemes have been worked out except the temperature control.” On May 8th Poole again wrote the Swindells: “We plan to make a general lay-out proposition first in the drawing room and then from these lay-outs, which will be approved *590 by your engineers, we will make detailed working drawings so as to proceed with the manufacture in the shop. We will be glad to have an order from you to proceed. * * *” In reply appellees in writing authorized Poole “to make up the detailed working drawings for the new Nixon machine, and to proceed with its manufacture when these drawings are mutually approved by your engineers and our machine department.” In the letter of April 30th Poole clearly recognized that the “job” was “experimental and development work,” and Davison in his letter to Shoemaker emphasizes that. . It is evident that the correspondence does not in itself and by itself constitute a complete contract. Poole’s letter of April 30th refers to “our plan this morning,” its letter of May 8th speaks of “lay-outs” to be approved by “your engineers,” and appellees’ order authorizes Poole to proceed only when the detailed working drawings “are mutually approved •by your engineers and our machine department.” So that the terms of the contract can only be known by reference to what took place in the negotiations concerning it between the parties prior to April 30th, to what they did and said in reference to it after May 11th, 1928, as well as to the correspondence. The correspondence itself, without such reference, furnishes no definite or intelligible guide as to what it was that Poole was to> do, other than to' make up certain preliminary drawings. There were at that time no specifications of the Nixon machine, nor were any completed drawings of it, as Poole proposed to modify and improve it, ever submitted to or approved by appellees as required by their order. And yet the work went on intermittently for over fourteen months before the Poole machine was. delivered, and during that period appellee paid Poole over $23,000 on account of it. Oertainly those payments were not made, nor was the work on the machine done, except under some contract, express or implied, the terms of which cannot be discovered in the correspondence itself. If reference is made to the oral negotiations between the parties, it is clear that the contract was ■either one under which Poole undertook to improve and strengthen the Nixon machine, so as to produce a durable *591 machine capable of manufacturing merchantable bottles at a rate substantially equal to that of the Nixon machine, or that it was one under which Poole undertook to furnish the material and mechanical skill needed to construct a machine in accordance with directions given it by appellees. Whether it was in the one form or the other depends upon whether the evidence of the appellant or the appellees is accepted, which of course was a jury question; but in no case can there be any intelligible definition of the contract except by reference to parol evidence as to the purpose and intentions of the parties, and under such circumstances reference to parol evidence to ascertain its terms is not a violation of the parol evidence rule but is necessary and proper. White Automobile Co. v. Dorsey, 119 Md. 268, 86 A. 617; Williston on Contracts, p. 1244. Considered together, the correspondence and the oral negotiations between the parties afford legally sufficient evidence that in consideration of the cost of construction plus twenty per cent. Poole undertook to construct for appellees a bottle-making machine of approximately the same dimensions as the Nixon machine, which would embody the Nixon moulding device, would be of sufficient strength and solidity to operate under ordinary working conditions, and would make merchantable bottles at a rate at least equal to that of the Nixon machine; that the two machines were to be substantially alike in design, except that the Poole machine was to be stronger than the Nixon machine and more substantial,, and was to embody an arrangement of the moulds different from that on the Nixon machine. The size of the projected machine was a material element of the contract, because it was a necessary factor in the use which appellees proposed to make of it, and of which they informed Poole when the contract was made. Through the failure of Poole to furnish plans, drawings,, specifications,, or other adequate information, appellees were unable to learn what the size of the Poole machine would be; until January, 1929, when they discovered that it would be too large for the use contemplated in the original contract. *592 That when they called the attention of Poole to that fact, its engineer assured them that its machine, while too large to be used in connection with another machine and a single feeder as originally intended, would make merchantable bottles at least twice as fast as the Nison machine, and so make the use of two machines unnecessary. That relying upon that assurance, the objection to the size of the Poole machine was waived, and Poole was authorized to proceed with the construction of the larger machine. Technically, such a contract was not one for work and material only, but a contract to sell “future goods” within the meaning of Oode, art. 83, secs. 22, 26, 35, 40, 85, 97. , As stated in Finney v. Apgar, 31 N. J. Law, 266, “There are few legal topics perplesed by a greater number of irreconcilable opinions than the question of law involved in the foregoing inquiry” as to whether such a contract as that made by the parties to this appeal is a contract for work, labor and material, or a contract to sell. And no doubt the confusion and uncertainty arising from that difficulty was one of the moving causes for the formation and adoption of the Uniform Sales Act in this and other states. The purpose of that act was to bring about greater uniformity and certainty in the law of sales throughout the United States by reducing its principles to clear, precise and comprehensive rules, and that purpose should not be thwarted by any narrow quibbling or falsely technical construction of its language. Section 5 of the act (Code, art. 83, sec. 26) declares that: “The goods which form the subject of a contract to sell may be * * * goods to be manufactured or acquired by the seller after the making of the contract to sell, in this act called '‘future goods.’ ” Section 76 of the Act (Oode, art. 83, sec. 97) provided that: “‘Goods’ include all chattels personal other than things in action or money. * * * ‘Euture goods’ means goods to' be manufactured or acquired by the seller after the making of the contract of sale.” That language would seem broad enough to cover any chattel to be manufactured and sold, but it is urged that it should be given a narrower meaning, more nearly in accord with the interpreta *593 tion given suc.li agreements in earlier cases in determining whether they were within the seventeenth section of the Statute of Frauds. In England it was held that such an agreement was a contract to sell (Lee v. Griffith, 1 Barn. & S. 272), but the rule prevailing in most American jurisdictions was otherwise, and that given in the Uniform Sales Act more nearly approximates what has been called the Massachusetts rule, which embodied as an exception to the English rule the principle that the Statute of Frauds did not apply where the “future goods” were to be manufactured especially for the buyer and were of such a character as not to be readily salable in the ordinary course of the seller’s business. That exception is found in the concluding paragraph of section 4, sub-section 2, of the Uniform Sales Act (Code, art. 83, sec. 25, subsec. 2), which reads as follows: “The provisions of this section apply to every such contract or sale, notwithstanding that the goods may be intended to be delivered at some future time, or may not at the time of such contract or sale be actually made, procured or provided, or fit or ready for delivery, or some act may be requisite for the making or completing thereof, or rendering the same fit for delivery; but if the goods are to be manufactured by the seller especially for the buyer and are not suitable for sale to others in the ordinary course of the seller’s business, the provisions of this section shall not apply.” In this state the rule as stated in such cases as Rentch v. Long, 27 Md. 197, that a contract for the delivery of goods where work and labor were to be bestowed upon them previous to delivery was not a contract to sell, was changed by the statute, and all “future goods” not within the exception were brought within the statute. Willard v. Higdon, 123 Md. 447, 91 A. 577; Stem v. Crawford, 133 Md. 579, 105 A. 780. And the same effect has been given the Uniform Sales Act in other states, although there is some apparent conflict in applying the exception to varying facts. Williston on Sales, secs. 54, 55, 55a; Benjamin on Sales, 177 et seq.; and cases collected in 1 Uniform Laws Ann., Uniform Sales Act, sec. 4, subsecs, b, c, d, e. *594 But section 4 of the act deals exclusively with the relation' of the Statute of Frauds to- the law of sales, and the language of the exception is expressly limited to the section in which it occurs, and does not, therefore, limit or qualify the language of any other section, although the fact that it was thought necessary does carry the implication that without it the section would have included “future goods,” whether made especially for the buyer, or whether suitable for sale in the ordinary course of the seller’s business. No such qualification is found in sections 5 or 76 of the act, and it must be assumed that those sections were intended to- cover sales of goods “to- be manufactured * * * by the seller after the making of the contract of sale,” whether such goods were or were not manufactured especially for the buyer, and whether they were or were not suitable for sale to others in the ordinary course of the seller’s business. In this case, assuming the truth of all evidence tending to support the appellees’ claim, the appellant, which appears to have been in the business of manufacturing machinery generally, undertook to make, for a price uncertain in its ultimate aggregate amount, but ascertainable through the- application of a fixed formula to- varying factors, a machine of a specified capacity and utility, which could, if it conformed to that specification, be sold commercially to persons other than the buyer. Such a contract was a contract to sell the machine within the meaning of sections 5 and 7 6 of the act (Code, art. 83, secs. 26, 97), and the evidence to which we have referred was legally sufficient to support that hypothesis, and also with one exception the other hypothesis of the- plaintiffs’ prayer. That exception was the hypothesis that appellant promised that the machine would make merchantable bottles in merchantable quantities. It may be inferred that it promised that the machine would make merchantable bottles, but there is no legally sufficient evidence- that appellant ever in iisdem verbis promised to make such bottles in “merchantable” quantities. As applied to the quality of the bottles, “merchantable” naturally and ordinarily would be accepted as meaning mar *595 ketable or salable, but as applied to tbe capacity of the machine it has no meaning. Oxford Dict. In itself and etymologically, the word has no necessary connection with quantity, and in the absence of some explanation its use was misleading, and the special exception to the prayer, on the ground that there was no evidence in the case legally sufficient to support the hypothesis that the defendant promised plaintiffs that the machine would make bottles in merchantable quantities, should have been sustained. The only evidence in the ease as to the rate of production was that the Poole machine would produce twice as many bottles as the Nixon machine, and the substitution of so indefinite a formula as “merchantable” for that fixed and definite standard was erroneous. The prayer, however, involves another and more substantial error. Although it goes to the plaintiffs’ right to recover, it entirely ignores evidence in the case tending to support the defense that the machine was made in conformity with a plan or design furnished by the appellees in accordance with their directions, and that its failure to make merchantable bottles at tire rate fixed by the contract was due to some error or fault in that plan, and not to any want of care or skill on the part of the appellant. That was a valid defense, and there was in the case evidence legally sufficient to support it. By failing to postulate it as a matter to be considered by the jury in determining whether the plaintiffs were entitled to recover, the prayer in effect withdrew that question from their consideration. For the jury could have found every fact recited in it, and yet the plaintiffs would not have been entitled to recover if the jury further found that the failure of the machine to function was due to some error or fault in its structural design, and that such design was determined by the plaintiffs, and not by the defendant. Adams v. Somerset County, 106 Md. 197, 66 A. 695; Mt. Vernon Brewing Co. v. Teschner, 108 Md. 174, 69 A. 702; Security Storage & Trust Co. v. Denys, 119 Md. 347, 86 A. 613; Howard County v. Pindell, 119 Md. 81, 85 A. 1041; Hart v. Leitch, 124 Md. 82, 91 A. 782; Harford County v. Harford County Sub. Imp. Assn., 134 Md. 552, 107 A. 348; Adkins v. Hast *596 ings, 138 Md. 465, 114 A. 288; Home Credit Co. v. Fouch, 155 Md. 397, 142 A. 515. Defendant’s fourth prayer, which was a general demurrer to the evidence, was, for reasons given supra, properly refused, as were its fifth prayer, which was a demurrer to' the evidence affecting the hypothesis of an implied warranty (Code, art. 83, sec. 36), its sixth prayer, which denied the legal sufficiency of the evidence to support a finding of an express warranty {Id., sec. 33), and its ninth prayer, which submitted the proposition that there was in the case m* legally sufficient evidence of an express warranty that the machine would make bottles at any particular rate or that it would make merchantable bottles at all. Its seventh prayer, while ' otherwise correct, was predicated upon the theory that if the plaintiffs suggested “changes in the design and construction” of the machine, that fact would defeat recovery, whether such changes were adopted or not. That theory was obviously ■untenable, and for that reason the prayer was properly rejected. The defendant’s eighth prayer submitted the proposition that if, at the inception of the contract, Poole was told that Nixon would supply all information “necessary in the construction of the machine regarding the manufacture of bottles,” and after that the parties agreed that Poole should construct a machine “embodying the ideas of the” Nixon machine and which would “substantially reproduce” it, all work to be subject to the approval of Nixon, there was no implied warranty. The vice of that prayer is that, although it is directed to a definite conclusion, it ignores evidence which if considered might lead to a different result from that directed. For while the conclusion directed by the prayer might be correct if the original contract, as supposed in the prayer, had never been modified or changed, it would not be correct if the original contract was subsequently changed, and under the substituted contract the defendant itself undertook to supply such skill, knowledge, labor and material as were needed to design as well as construct a machine capable of producing the specified result. There *597 was in the case evidence legally sufficient to permit such an inference, and, as the effect of the prayer was to withdraw that evidence from the consideration of the jury, it was properly refused. Eor the errors involved in granting the plaintiffs’ prayer, the judgment appealed from must be reversed, and the case remanded for a new trial. Judgment reversed, and case remanded for a new trial, with costs to the appellant.