ABBOTT MACHINE CO. v. UNIVERSAL WINDING CO.

No. 3832.Circuit Court of Appeals, First Circuit.
July 6, 1943.

Appeal from the District Court of the United States for the District of New Hampshire; John A. Peters, Judge.

Action by the Universal Winding Company against the Abbott Machine Company for a declaratory judgment that certain patent claims are invalid or not infringed by plaintiff, in which defendant filed a counterclaim for infringement of the patent. Judgment for plaintiff, D.C.N.H., 43 F. Supp. 647, and defendant appeals.

Affirmed.

Robert Cushman and Roberts, Cushman Woodberry, all of Boston, Mass., for appellant.

George P. Dike, Cedric W. Porter, George P. Towle, Jr., and Dike, Calver Porter, all of Boston, Mass., for appellee.

Before MAGRUDER, MAHONEY, and WOODBURY, Circuit Judges.

WOODBURY, Circuit Judge.

The plaintiff-appellee, Universal Winding Company, a Massachusetts corporation, filed a petition for a declaratory judgment in the court below against the defendant-appellant, Abbott Machine Company, a New Hampshire corporation, asking for an adjudication that fourteen of the forty-one claims of patent number 2,160,810, issued on June 6, 1939, to Edward J. Abbott and assigned by him to the defendant are invalid, or, if valid, are not infringed by certain machines manufactured by the plaintiff. The defendant answered and filed a counterclaim alleging the validity of its patent and infringement thereof by the plaintiff. At the trial in the district court the defendant placed little if any reliance upon twelve of the fourteen claims in issue and subsequently disclaimed as to them and some of the others not in issue. The plaintiff eventually admitted that if the claims mentioned in its complaint are valid its machines infringe. Thus the issue is narrowed to the validity of only two claims, — the 14th and 28th. The court below found both of them invalid and entered judgment accordingly. The defendant thereupon took this appeal to us.

The Abbott patent is for a fully automatic machine for winding what are called filling wound bobbins, that is, bobbins for use in the shuttles of looms. These filling wound bobbins are held by their butt ends in the cavities of shuttles and the yarn wound upon them is paid out through an eye in the end of the shuttle near the tip of the bobbin. Due to the limited space within a shuttle for the accommodation of a bobbin and the great speed, particularly in a modern automatic loom, with which the yarn must be delivered as the shuttle is shot back and forth through the shed formed by the alternate raising and lowering of the warp threads, filling wound bobbins have to be wound in a particular way so the yarn will come off of them with great rapidity but without tangling or breaking. The district court in its memorandum opinion said that this special form of winding “is accomplished by laying the thread on the bobbin in a series of layers like nested cones[1] , each cone-shaped

Page 167

layer being wound upon the next preceding one and being axially displaced slightly toward the tip of the bobbin as winding proceeds. The finished thread mass has a cylindrical shape throughout the greater part of its length, terminating toward the tip of the bobbin in a tapered or cone-shaped surface. The cylindrical part of the surface, however, is formed not by cylindrical layers of thread, one over the other, but by the bases only of the several nested cones of wound thread. By this method of winding the thread on the bobbin the growth of the thread mass as the winding proceeds is in a lengthwise or axial direction from the butt to the tip end of the bobbin. This method of winding and form of package is necessary so that the thread may be unwound by being drawn lengthwise from the tip of the bobbin; that is, always from the last cone-shaped layer of thread nearest the tip, so that the thread in coming off does not have to drag across the length of the package, as it would do if the layers of thread were laid cylindrically throughout the whole length of the bobbin.”

Filling wound bobbins have been wound in this way for a great many years and they have been so wound by machinery, so far as the present record indicates, at least since the issuance of a patent to Albert Ball in 1887 for what is called a “cop-building mechanism for ring-spinning frames”. But Ball’s machine and others which followed it were not automatic, although automatic machinery for winding other kinds of packages of yarns have been in use for more than eighty years. The district court did not find the reason for this delay in developing automatic machines for winding filling wound bobbins but only said: “for some reason, not perfectly clear, automatic features were not applied to the winding of filling bobbins until the time of the Abbott invention.”

The plaintiff takes the position, and its position was sustained by the court below, that in view of the automatic machines of the prior art for winding many kinds of thread packages, particularly spools of thread for domestic use, it was only the every day work of a machine designer to put together an automatic machine to do any kind of winding, including winding filling wound bobbins. The defendant, on the other hand, contends that since no single prior art machine contained the combination of parts, which it admits are old, covered by the claims of the Abbott patent here in issue, and since this new combination of old parts resulted in a machine of vastly increased productive capacity (from 250 bobbins per hour to 1000 to 2000 bobbins per hour) Abbott is entitled as a matter of law to rank as an inventor of outstanding ability. And it says that its view is reinforced, if not made well nigh inescapable, by the undisputed fact that the art waited nearly eighty years for the advent of a machine such as Abbott’s. The question, then, is whether the claims in suit, which are copied in the margin,[2] cover a patentable combination of parts or elements of machines of the prior art, or whether they cover only what has been called a “mere” or “non-inventive” aggregation or assembly of those parts.

It has long been recognized by the Supreme Court that some combinations of old mechanical elements or parts are patentable while others are not. Hailes v. Van Wormer, 20 Wall. 353, 368, 22 L.Ed. 241. Those which are patentable have been called “true” or “patentable” combinations; those which are not, as we

Page 168

have indicated above, have been called “aggregations”, usually preceded by some belittling adjective. Obviously terminology cannot decide cases, and the test to determine whether or not any given combination or aggregation — the terms are really synonymous — was stated in the case just cited as follows: “It must be conceded that a new combination, if it produces new and useful results, is patentable, though all the constituents of the combination were well known and in common use before the combination was made. But the results must be a product of the combination, and not a mere aggregate of several results each the complete product of one of the combined elements. Combined results are not necessarily a novel result, nor are they an old result obtained in a new and improved manner. Merely bringing old devices into juxtaposition, and there allowing each to work out its own effect without the production of something novel, is not invention. No one by bringing together several old devices without producing a new and useful result the joint product of the elements of the combination and something more than an aggregate of old results, can acquire a right to prevent others from using the same devices, either singly or in other combinations, or, even if a new and useful result is obtained, can prevent others from using some of the devices, omitting others, in combination.”

In later cases the Supreme Court reaffirmed and elaborated upon the test stated in the quotation above. Thus in Reckendorfer v. Faber, 92 U.S. 347, 357, 23 L.Ed. 719, the Supreme Court, in denying patentability to the combination of a lead pencil and an eraser, said: “The combination, to be patentable, must produce a different force or effect, or result in the combined forces or processes, from that given by their separate parts. There must be a new result produced by their union: if not so, it is only an aggregation of separate elements.” And later, in Pickering v. McCullough, 104 U.S. 310, 318, 26 L.Ed. 749, the court said: “In a patentable combination of old elements, all the constituents must so enter into it as that each qualifies every other * * *. It must form either a new machine of a distinct character and function, or produce a result due to the joint and co-operating action of all the elements, and which is not the mere adding together of separate contributions. Otherwise it is only a mechanical juxtaposition, and not a vital union.” This test, however, was stated in more guarded language in Loom Co. v. Higgins, 105 U.S. 580, 591, 26 L.Ed. 1177, when the court said: “It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention.” See also Knapp v. Morss, 150 U.S. 221, 227, 14 S.Ct. 81, 37 L.Ed. 1059, and Richards v. Chase Elevator Co., 158 U.S. 299, 302, 15 S.Ct. 831, 39 L.Ed. 991.

But the statements in these cases do not mean, as the plaintiff argues, that there can only be a patentable combination when the old elements are combined in such a way that each modifies the actions of the others so that all cooperate in a new way to produce a new result. The Circuit Court of Appeals, Third Circuit, in National Cash Register Co. v. American Cash Register Co., 53 F. 367, exploded this argument by pointing out that each mechanical element in a combination must of necessity act “according to the law of its own being”, (53 F. at page 371) — that, for instance, (53 F. at page 372) “A screw or a lever can act only in one way” and that in combination neither can make the other act in any different way, it being enough for patentability under the rule stated above if they “so act in combination as to produce, in consequence of their combination, a single, new, and useful result.”

In recent years, however, the Supreme Court in testing the patentability of mechanical combinations seems to have looked less to the joint as distinguished from the separate and independent functioning of the mechanical elements in producing the new and useful result essential to patentability under the statute, and more to the broader question of whether or not the inventive faculty rather than the mere skill of the calling was required to envisage the combination covered by the patent in suit. Thus in Concrete Appliances Co. v. Gomery, 269 U.S. 177, 185, 46 S.Ct. 42, 45, 70 L.Ed. 222, the Supreme Court, in holding invalid a patent for an apparatus for distributing “wet” or “mush” concrete from the mixer to the places where it is to be incorporated into a building in process of construction, said: “The adaptation independently made by engineers and builders

Page 169

of these familiar appliances to the movement and distribution of concrete cement in building operations and the independent patent applications within a comparatively short space of time, for devices for that purpose are in themselves persuasive evidence that this use in combination of well known mechanical elements was the product only of ordinary mechanical or engineering skill and not of inventive genius.” And in Toledo Pressed Steel Co. v. Standard Parts, Inc., 307 U.S. 350, 356, 59 S.Ct. 897, 899, 83 L.Ed. 1334, the same court, in striking down a patent for a kerosene flare for outdoor use as a warning signal which had a shielding device for protecting it from wind and rain, said: “On the records before us, it is impossible to hold that production of the patented device required more than mechanical skill and originality attributable to those familiar with the art of protecting flames of kerosene and other burners.”

This does not mean that the test of cooperation between parts has been abandoned. It has been mentioned and relied upon, to some extent at least, in several recent cases, including the case cited last above. See also Powers-Kennedy Co. v. Concrete Co., 282 U.S. 175, 186, 51 S.Ct. 95, 75 L.Ed. 278; Altoona Publix Theatres, Inc., v. Tri-Ergon Corp., 294 U.S. 477, 486, 55 S.Ct. 455, 79 L.Ed. 1005. We only mean that now-a-days the emphasis in cases of this sort is less upon an analysis of the machine to see whether or not its parts cooperate to produce some new and useful result, and more upon an analysis of the machine and of the prior art to see whether or not the faculty of invention was required to put its parts together in such a way as to accomplish that new and useful result. This is illustrated by the recent case of Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 62 S.Ct. 37, 86 L.Ed. 58. In this case there is no discussion of the question of whether or not the parts assembled cooperate with one another, the court holding simply that although a new and useful result was achieved, it was not achieved by an exercise of the inventive faculty in putting the combination together. The court said, (314 U.S. at page 88, 62 S.Ct. at page 39, 86 L.Ed. 58) citing the Altoona Publix Theatres case supra, that in its opinion “the Mead device was not the result of invention but a `mere exercise of the skill of the calling’, an advance `plainly indicated by the prior art'”; that the question before it in the case was “whether it was invention for one skilled in the art and familiar with Morris and Copeland, and with the extensive use of the automatic thermostatic control of an electric heating circuit, to apply the Copeland automatic circuit to the Morris removable heating unit in substitution for a circuit manually controlled”, and it cited Hailes v. Van Wormer, and the cases which followed it, not in support of the test of cooperation of parts, but for the proposition that “More must be done than to utilize the skill of the art in bringing old tools into new combinations.”

Whatever the reason for this change in emphasis may be, it seems to us that the cooperation test for determining the patentability of a combination established in Hailes v. Van Wormer, although adequate in many cases, is not adequate in all. It was adequate to determine the patentability of the lead pencil and eraser combination under consideration in the case of Reckendorfer v. Faber, 92 U.S. 347, 23 L.Ed. 719, because clearly there was no cooperation of function between the lead on one end of a pencil and the eraser on the other, and, in addition, at least by modern standards, one could hardly say that an exercise of the faculty of invention was required to engender the idea of combining them. But the test of cooperation between parts breaks down when applied to the combination under consideration in the Altoona Publix Theatres v. Tri-Ergon Corp. case. The patentee in that case, to give uniformity of speed to a device for the projection of talking moving pictures, added a flywheel to a previously known mechanism for moving the strip of film in the projector. Clearly the flywheel cooperated with the other parts of the mechanism by imparting to them uniformity and consistency of speed, but the court (294 U.S. at page 483, 55 S.Ct. at page 457, 79 L.Ed. 1005), noting that a “flywheel set upon a revolving shaft is an ancient mechanical device for securing continuity and uniformity of motion when brought into association with any form of machinery moved by intermittent force or meeting with irregular or intermittent resistance”, held the patent invalid on the ground that the combination was (294 U.S.

Page 170

at page 486, 55 S.Ct. at page 459, 79 L.Ed. 1005) “the product of skill, not of invention.”

In the case at bar the district judge held that the claims in issue failed to meet either test. He said in his memorandum opinion [43 F. Supp. 647, 651]: “Here I can see no joint function performed which produces a new result or any old result in any other way than it was formerly accomplished”, and “In the case before me it is difficult to see how an art which knew how to wind packages of yarns automatically and knew how to wind filling packages by hand, required some uncommon talent merely to conceive of combining the two.”

In keeping with the trend of the modern decisions, particularly the Cuno case, we shall not attempt to determine whether or not the parts of the defendant’s mechanism cooperate with one another to produce a new and useful result, but, conceding that a new and useful result is achieved by the combination,[3] we shall determine whether or not the court below was correct in concluding that no “uncommon talent” was required to put it together.[4]

We turn now to a consideration of the patented machine and of the machines of the prior art.

The district court in its memorandum opinion said: “The essential parts of a winding-machine consist of spindle centers for grasping and rotating[5] the bobbin on which the thread is wound; a thread-guide which lays each turn of the thread in the right position on the revolving bobbin, and some means such as a cam to move the thread-guide to and fro lengthwise of the bobbin. The shape of the finished package depends upon the movement of the thread-guide back and forth on the bobbin.” This brief statement of general principle of operation is elaborated and explained in relation to the plaintiff’s and defendant’s machines in the findings of fact as follows:

“All kinds of packages, such as filling wound bobbins, spools, cops, etc., are wound by the relative movement of a spindle or a core and a thread guide. In some machines the spindle or core is moved axially and the thread guide or flier is moved around the spindle or core, but in some cases the spindle or core is rotated and the thread guide reciprocated axially of the core during winding.

“In machines where the spindle revolves and the thread guide moves axially,[6] the relative speed of rotation of the core and of reciprocation of the thread guide, the extent of reciprocation, and the location of the zone of the reciprocation or traverse determine the shape and character of the package produced, i.e., the shape of the finished package depends upon the movement of the thread guide back and forth on the bobbin. The axial movement of the thread guide is called `traversing’; the distance which the thread guide moves before reversing its movement is called the `stroke’, and the place or zone where the movement takes place is called the `zone of traverse’. If the thread guide has a uniform reciprocation in the same location, a cylindrical package with flat ends will result, except when the thread is wound on a conical core, in which case a cone results. When a spool with sloping flanges is to be wound, the length of traverse of each reciprocation is increased slightly at each end of the stroke. When a filling wound bobbin is wound the length of the stroke of the thread guide remains the same and is relatively short compared to the length of the package, and the zone of traverse is moved progressively toward the tip of the bobbin.”

From these statements it is apparent that there are two principal parts or assemblies

Page 171

in any winding machine. There is the assembly which grasps the empty bobbin and rotates it while it is being wound with yarn or thread; and there is the assembly which controls the way in which the thread is wound upon the bobbin. These for convenience may be called the rotating mechanism and the thread-guide mechanism. In the Abbott machine both of these mechanisms are fully automatic so that, barring breakdowns, all the operator of the machine is called upon to do is to see that the machine is supplied with yarn for winding, that the magazine of the machine is constantly supplied with empty bobbins, and that the container into which full bobbins are dropped is removed when it is full and replaced by an empty one. Substantially the same may be said with respect to the duties of operators of the plaintiff’s infringing machines.

Further description of the Abbott machine may conveniently be postponed until later and we will turn to the machines of the prior art.

For many years before the advent of Abbott’s automatic machine the standard machine used for winding filling wound bobbins was a manually-tended one manufactured by the plaintiff called Universal No. 90. This machine consisted of a series or battery of winding units each fed with yarn from a separate source of supply. Each unit consisted of a rotating spindle for turning the bobbin and a traversing thread guide mechanism for laying the yarn on the bobbin. The duties of the operator of this machine were to wind the end of yarn hanging from the thread guide around the spindle and then to impale an empty bobbin on the spindle, thus pinching the end of the yarn between the spindle and the hole through the axis of the bobbin so that when the spindle with its bobbin was rotated winding would begin. The next operation was to start the unit by pushing a lever. This caused the spindle with its bobbin to revolve and the thread guide to reciprocate back and forth through a zone of traverse of an inch and a half or so. By a mechanism which we need not describe, the thread-guide mechanism was made to advance the zone of traverse axially toward the tip of the bobbin so that the yarn was laid in cone shaped layers. When the bobbin was fully wound the unit automatically stopped. The operator then manually pushed the thread-guide mechanism back so that it was in position to start winding on the butt end of the next bobbin, pulled the full bobbin off the spindle, broke or cut the yarn between it and the thread-guide, and then pushed a new bobbin onto the spindle as already indicated, thus completing the cycle of operations.

Abbott’s thread-guide mechanism and that of the Universal No. 90 machine are very similar except that Abbott’s is made to return to starting position automatically as soon as a bobbin is fully wound. This feature of Abbott’s machine, however, was not new with him. It is shown in a patent (No. 1,236,906) issued to Calkins on August 14, 1917, as well as in some others.

Abbott’s bobbin clamping and rotating mechanism, however, differs radically from that used in the No. 90 machine. In Abbott’s machine the bobbin is not held on a spindle but is clamped by its ends between two retractable, cam-operated, cup-shaped grips called spindle centers or end clamps, the outer one holding the tip of the bobbin and the inner one its butt end. The inner end clamp is rotating and serves both to rotate the bobbin and to grip the end of yarn hanging from the thread-guide by pinching it against the butt of the bobbin so that when rotation begins the yarn will be wound on the bobbin. The outer end clamp is rotatable and serves, with the inner one, to hold the bobbin firmly in position while it is being wound. The defendant in its brief describes the operation of Abbott’s machine as follows:

“The winding and traversing are both stopped when the traverse mechanism arrives at the tip end of the full bobbin. The bobbin with the filling-wound yarn on it is then dropped out from between the end clamping members by the action of a cam which retracts one of the end clamps. Another cam pushes the traverse mechanism back to the position at the beginning end of the bobbin and the yarn is also carried over from the remote or final end of the full bobbin to the position at the beginning end of a new bobbin and is held across the inner bobbin clamp in time to be there when a fresh bobbin is supplied from a magazine of empty bobbins and grasped at each end by the bobbin clamps. The yarn is then clamped between the innermost bobbin clamp and one end of the new bobbin. The thread is then cut off between the new empty

Page 172

bobbin and the filled bobbin just discharged. The winding and traversing are again started in the beginning zone of the new bobbin and the cycle is repeated.”

Abbott’s conception of picking up, holding and rotating the bobbin, catching the yarn as it extends from the thread-guide between one end of the empty bobbin about to be wound and a spindle center so that when rotating begins winding will take place; cutting the yarn after a bobbin has been wound and is ready to be dropped from the machine, and dropping the full bobbin was not original with him. These features all appear in machines for winding spools of thread for domestic use which have an English patent history extending back to 1858, some of which were in use in Groton, Connecticut, more than two years prior to Abbott’s application, and they also appear in a winding machine described in a patent (No. 482,308) issued to Hill in 1892 and in one described in a patent (No. 1,121,103) issued to Hooper in 1914.

From the foregoing it is apparent that what Abbott did was to combine two devices of the prior art — an automatic core holding and rotating device which had been used in machines for winding spools of thread for domestic use, but which could readily be re-designed to hold bobbins, and an automatic thread-guide mechanism which had been used in non-automatic machines for winding filling wound bobbins. And this combination, as we have already indicated, produced a new and useful result. So we can say, as the Supreme Court said in Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 90, 62 S.Ct. 37, 40, 86 L.Ed. 58, “We may concede that the functions performed by Mead’s [in the case at bar, Abbott’s] combination were new and useful. But that does not necessarily make the device patentable. Under the statute, 35 U.S.C. § 31, 35 U.S.C.A. § 31, R.S. § 4886 the device must not only be `new and useful’, it must also be an `invention’ or `discovery’. * * * Since * * * 1851, it has been recognized that if an improvement is to obtain the privileged position of a patent more ingenuity must be involved than the work of a mechanic skilled in the art.”

The district court, as we have already pointed out, concluded that Abbott, in conceiving his combination, exercised only the skill of his calling as indicated by the prior art. We are of the view that this conclusion is not only not clearly erroneous but is clearly correct. Considering the established place in the prior art of both of the mechanical elements combined by Abbott, and the obvious increase in production which would result from making his combination, it seems to us evident that the combination covered by the claims in issue is one that would naturally and normally occur to anyone working in the field. The combination made by Abbott seems to us comparable to the combinations made by the patentees in the Cuno case, in the case of Textile Machine Works v. Hirsch Textile Machines, Inc., 302 U.S. 490, 58 S.Ct. 291, 82 L.Ed. 382, which the case at bar closely resembles, as well as in several of the other recent Supreme Court cases which we have cited. In view of these cases we cannot call Abbott’s step that of an inventor but must characterize it only as an advance expectable from one skilled in the art.

There remains to be considered the defendant’s argument that if Abbott’s step was so obvious, why did the art wait nearly eighty years for a fully automatic machine for winding filling wound bobbins?

The plaintiff answers this question by saying that during those eighty or so years there was no need for an automatic machine such as Abbott’s and consequently no demand for one. It says that the reason for this is that until the last two decades wages in the textile industry were low and hence labor saving machines were not a crying need of that industry, and also that rewinding filling wound bobbins, the use to which fully automatic machines are usually put, was not of any great importance until the advent, also during the last two decades, of high speed automatic looms. The plaintiff then counters by asking the defendant why, if there was a need for an automatic machine, was only one Abbott machine sold from the date of his application for a patent on November 24, 1931, to the date of trial some ten years later?

We incline to the plaintiff’s side of this argument because it seems to us that, the combination being an obvious one, the industry would have overcome its inertia and taken up what is apparently a labor saving machine if there had been any demand for it. But however this may be, we do not have to, nor can we from the

Page 173

findings of the district court, give a definite answer to the opposing questions. What we have said sufficiently indicates that the argument that invention is shown by lapse of time is so weakened that it cannot tip the scales in favor of patentability. And our position is supported by the conclusion of the district court that: “On the question of invention also it should not be overlooked that when it apparently became desirable to convert manual operation into mechanical operation several independent inventions of the same thing appeared substantially simultaneously. That is, the Abbott machine, the Universal and Schweiter Whitin’s, as shown by the evidence.”

The judgment of the District Court is affirmed with costs to the appellee.

[1] Counsel for the plaintiff described these layers as “like a series of flowerpots, stuck on top of another.”
[2] “Claim 14. A winding machine having therein means for winding a strand from a supply onto a rotating core, means for traversing the strand on the core, means for causing a progressive shifting of the range of traverse in a direction along the axis of the core during winding, in combination with means for automatically stopping such traversing upon completion of the package, means for returning the traverse means to the beginning end of a core, means for automatically associating the strand with a fresh core for winding thereon, and means for automatically resuming traversing of the strand.

“Claim 28. A winding machine having therein core end-clamping mechanism acting automatically to clamp a core for winding, means for automatically supplying a core to said core end-clamping mechanism, and means operable after clamping of the core for rotating the clamped core to wind strand thereon, in combination with a traversing strand guide adapted to distribute the winding strand on the rotating core, the strand guide having a full stroke less than the length of the package, means for shifting the zone of traverse stroke of the guide axially of the core during winding, and means for automatically causing the zone of the strand guide stroke to lie at one end of the core at the start of winding thereon.”

[3] Abbott very substantially increased the productivity of machines for winding filling wound bobbins and increasing the productivity of a machine has for a long time been considered a new and useful result under the statute. Loom Company v. Higgins, 105 U.S. 580, 591, 592, 26 L.Ed. 1177; Eibel Process Co. v. Minnesota Ontario Paper Co., 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523.
[4] We agree with Judge Learned Hand’s statement in the recent case of Weston Electrical Instrument Corp. v. Dejur-Amsco Corp., 2 Cir., 133 F.2d 778, 780, 781, that “In every case the issue is whether a person acquainted with what had already been done, and endowed with no more than average imagination would have conceived the combination; and, while so to state the problem is not much of a step towards its answer, at least it clears the path of false directions.”
[5] In some winding machines the bobbin is held and rotated by being impaled on a tapered rotating rod or spindle.
[6] Both the plaintiff’s and defendant’s machines operate in this way.

Page 627