STONE, Circuit Judge.

This is an action for infringement of a patent to W. K. Howe, No. 1,551,515, filed November 12, 1919, issued August 25, 1925, for an automatic train control device. The plaintiff, assignee of Howe, makes and sells a train control device following the Howe patent. The defendant is a railway company which has installed and uses a train control *791device made by the Sprague Safety Control & Signal Company. Without specifically determining the validity of the patent, the trial court held “that if the Howe patent involved invention at all, the claims can receive only a very narrow construction, and that Howe is practically limited to the exact structure described by him. * * * I do not hold the patent invalid, but I do hold that there was no infringement.”

The defenses were lack of inventive novelty, anticipation, and no infringement. The patent in suit is a combination patent. Because of the character of the patent, of the defenses urged, and of the view taken by the trial court, it is necessary to examine the field of art in which these devices belong. There are various kinds of train control devices differing in respects which admit of certain classifications. The devices here involved are of the “intermittent inductive noncontact type” based on the block signal system. The particular devices here involved and the citations claimed to be anticipatory can be best understood through a logical process of sketching the entire art and narrowing down to that section of the art in which these devices lie.

General Sketch of the Art.

The safety of passengers and employees on railway trains has long been a serious problem. Several means of promoting such safety have proven successful. Among these are the air brake and the block signal system. The main purpose and effect of the air brake is to furnish a powerful braking energy upon all the wheels of the train through a single control device operable from the engine so that the speed of the train can be readily and effectively controlled or the train quickly stopped when danger becomes apparent to the engineer. The main purpose and effect of the block signal system is to give the engineer warning of danger before it might otherwise be apparent to him. The air brake is universally used on American railways and the block signal system is in general use.

Even with the block signal system, experience demonstrated that the engineer sometimes failed to notice or to properly interpret such warning or refused to beed it, thus causing accidents which resulted in death or injury.1 To nullify or minimize the effect of these defects in the human element involved, devices were invented which were designed to protect the train by automatically slowing and/or stopping it as it approached a danger condition. These are known as automatic train control devices. The fundamental purpose of any train control device is, where a dangerous condition exists ahead of the train, to control the speed of the train automatically when the engineer for any reason fails to take proper action. This control naturally took the form of slackening the speed, of slackening the speed with ultimate stoppage of the train or of stoppage of the train as quickly as possible.

Some devices operated independently of any block signal system; others were adapted to act independently or in conjunction with a block system; still others were based upon connection with such system. The devices here are based upon the block system.

Some devices were designed to act only at fixed stations or locations on the track, and were called “intermittent” types; others were constantly subject to action, and were called “continuous” types. The devices here .axe “intermittent.”

■ While all types made some character of use of the compressed air of the air brake system, some were otherwise purely mechanical, while some were electric or electromagnetic. The latter were put in operation by electromagnetic impulses received from the track or from track units inductively, and therefore were known as “inductive” types. The devices here are of the “inductive” type.

Some secured the danger impulse from the track element to the engine element through physical contact between such ele-mente; some had no such contact, and were known as “noncontact” types. The devices here are “noncontact” types.

While all train control devices had the same broad general purpose, the above enumeration of recognized classes of such de-viees suggests that considerable difference existed in the ideas as to the lines along which the general purpose should be worked out, and also that such difference would result in a variety of devices. In spite of or as a result of such differences, there developed certain conceptions as to what were necessary elements and as to what were desirable elements in any train control device. As the automatic control was to be exercised on the air brakes, which were controlled on the en-’ gine, only when danger was ahead, it was evident that such control must he from an engine installed element which would actuate the *792brake system, and that the engine element must be set in motion by some impulse from a track element which would indicate danger ahead. Also, when the brakes were automatically set, there must be some means to release them and to reset tbe device so that it would be subject to the next danger indication. These elements were necessary to the effective operation of any train control device.

Another element was deemed highly desirable. This element bad to do with tbe control of the engineer over the automatic braking. Without some such control, the automatic device would apply the brakes even though the engineer were fully alive and obedient to the situation and no matter what the' attendant circumstances. The stopping and starting of trains is attended with expense and operating delay. Under some conditions, automatic stopping was unnecessary; under others, undesirable; under others, dangerous. The railroads contended that it was very desirable to avoid this inescapable stopping if the engineer were alert, because bis trained judgment could be relied. upon to avert tbe danger and to take proper action under all the circumstances present. There Was a decided difference among inventors and others 2 as to whether any such control should be given the engineer. Tbe final conclusion Was in favor of sueb control, and. it is now *793recognized as a highly desirable element of an automatic train control system. Such permissive elements are variously, and somewhat interchangeably, called “hold off,” “forestall? ing,” or “annulling” devices (exactness might require hold-off and forestalling to apply to ■action to prevent automatic braking before .it starts, while annulling would apply to nullifying a braking operation after it has start•ed). The practical working out of- this element naturally gave rise to different theories :and resulting different devices. These theories arose from the various views as to the •extent of such permissive control and as to the conditions and results of its exercise deemed, by the particular inventor, most consistent with safety and with facility of train operation, and also as to character of penalties to be inflicted on the engineer. It is important to keep in mind the existence of -such differences in theory, because, while a theory is not patentable, yet the purpose of a device may have a bearing upon its novelty or its scope or upon its dissimilarity to other ■devices having the same or somewhat different purposes.

■While the two necessary elements of (1) a train control element, made up of track and ■engine features, and (2) a resetting element ■combined with (3) a desirable hold-off element, were sufficient to comprehend a well-rounded automatic train control device, there was yet another feature developed which was deemed advisable by some. That feature was that the entire device should be protected from manipulation which would prevent it functioning in the manner intended. Thus was. reached the final general conception of a complete train control system as being one including (1) automatic control, (2) reset, (3) hold-off, and (4) self-protection from improper manipulation. Naturally, this complete conception was reached piecemeal and only through the thought and efforts of many men, principally inventors, railway men, and the Interstate Commerce Commission organization,3 working alone or together. It was an advance with little chronology or regularity of progress.

For the purposes of this case, it is sufficient to know that every one of these four elements were known to and represented in the art when Howe entered the field.

Howe makes no claim to be a pioneer in automatic train control nor of such control by the intermittent inductive noncontact type based upon a block signal system. He makes no such claim as to any one of the four above elements. His is a combination patent which includes all of the four elements. His contention is that he came into the field with “a new plan of train control” which was realized through a novel device. Since Howe’s device is based on the block signal and is intended to work as a connection between the existing air brake .system on the train and the existing block signal system on the track, it is necessary to have an outline conception of those systems in order to appreciate the purposes of the inventor and particularly, to understand the method he designed to work out those purposes.

A train air brake system employs compressed air to control application of brakes on the engine and the separate cars of a train. Air, which is compressed on the engine and stored in reservoirs thereon, is carried to brake control mechanism on each car ■and on the engine. Relative maintenance of pressure at a predetermined rate holds the brakes free from the wheels. The brakes axe applied in ratio with the reduction of this pressure. The means of reducing this pressure is by releasing the compressed air to atmosphere through suitably located valve *794vents. The control of these vents is centered in an engine valve operated by a small hand lever within easy reach of the engineer. This valve has a position marked dial plate to guide the engineer in operation. The marked positions are, successively from left to right, “release,” “running,” “holding,” “lap,” “service,” and “emergency.” The three first perform much the same functions, and may be generalized as the release condition, when no braking force is being exerted on the wheels —this is the normal running condition, and also is where the brake valve handle is moved to release the brakes. The lap condition is where the whole brake equipment is maintained in the position it was just previously. The service position is that in which the brake valve is brought to apply the brakes in the usual normal manner. The emergency position is a severe application of the brakes designed to stop the train as quickly as possible. Service and emergency positions are the brake-applying positions, and differ principally in the intensity of the application.

Since any brake application is brought about by venting the air in the service line, one problem of the automatic train control inventor was to have his device work to vent the brake system air. This he might do by vents entirely independent of those controlled by the regular engine valve; by the regular vents controlled through the regular brake valve or controlled independent thereof; or by some combination of the above.

The bloek system employs" electrically operated wayside signals to indicate the safety condition of definite lengths of track. Each length is ealled a “bloek,” and is usually (not necessarily) a mile of track. At each end of every block is a signal station capable of showing three signals — clear, caution, or danger. The signals are controlled by electric circuits which include the rails in each bloek. Normally these circuits are closed and the signals are at “clear.” When the circuit of a given bloek is opened, as by a broken rail or open switch or by a car or train thereon, the signal at each end of that block goes to “danger” position. Also the signal at the far or “distant” end of each of the bloeks adjoining the “danger” block, go to “caution.” Thus the engineer is warned of danger, by the “caution,” a complete bloek before reaching that in which the danger exists. The purpose of the “caution” is to provide warning of danger while there is still time and distance to reduce the running speed so that the train may, if necessary, be stopped at the entrance to the “danger” block.

To utilize this system, the automatic control inventor must provide means by which the warning from the block signals could automatically be transmitted and registered into braking action on the train.

Having sketched the broad field of the art into which Howe entered, and having indicated the class in that field to which Howe’s device belongs, we pass to a particular examination of his patent, and thereafter of the cited art, with a view of determining the novelty of his patent and the scope to be given the litigated claims thereof.

Howe’s Patent.

Howe had a.main conception or purpose which ran through his devices, in this patent, and which governed the form, component elements, and functioning thereof. That conception was to encourage the engineer to watch for and acknowledge every block signal warning by making it easy for him to do so, and by inflicting a prescribed penalty for failure so to do. The character of the “penalty” he deemed sufficient resulted in two forms of engine devices which are alike in some respects and different in others. Either of these two forms have ranges of adaptability, but, as this adaptability depends mainly upon the location of the track elements and as the track elements are not involved here,, it is necessary only to note that such exist.' One of the two types of engine devices, shown in Fig. 2 of the patent, is not involved here.4 •The form here involved is that shown by Fig.. 5 of the patent. The commercial installa*795fcion more nearly followed the form (Fig. 5) in suit. We will confine ourselves to the form in suit, referring to other features only oeeasionally.

The fundamental conception in the patent (Fig. 5) is a single simple brake-setting unit operated by one electric circuit, which eir-cuit is entirely beyond the control of the *796engineer except through the resetting and the hold-off devices and through those devices only under definitely limited and inescapable conditions insuring proper operation of the reset and insuring alert affirmative acknowledgment of the block signals by the engineer as to the hold-off. The resetting and the hold-off devices are effective only as' either of them affects or controls this circuit. The control over this circuit, either by the reset or by the hold-off, is definitely and narrowly limited; such precise limitations are automatically enforced, and such limitations cannot he exceeded, because such enforcement cannot be prevented — thus one result of the mechanism is to insure its own automatic operation unless such operation be prevented under the precise conditions intentionally provided for in that mechanism.

Passing from this general statement of the device, it can be best understood in detail by separately describing the several elfe-ments thereof and the relation of each to that entirety.

Brake-Setting Unit.

The principle of this unit is to utilize the force of compressed air to initiate and compel the operation of the ordinary air brake system. This action is directly on the engine air brake valve. A sector 41 is fastened to the engine brake valve stem E so that, when the sector is swung, the valve will shift to apply the brakes.5 This sector meshes with a rack 42, which is integral or attached to a rod 43. This rod acts as a piston rod for an operating cylinder 44, and passes through a piston 45 loosely so that when the automatic device is inactive the manual operation of the brakes is uninterfered with. Adjustable nuts 46 on the rod and stops 47 in the cylinder provide means of adjustment to regulate the exact degree of automatic braking desired. The piston is automatically actuated by the introduction, near the left end thereof, of compressed air stored in a main reservoir M. R. The passage of this air from the reservoir to the cylinder is controlled by an interposed electyopneumatie valve. This valve had two chambers — an upper 106, connected with the reservoir M. R. by a pipe 107, and a lower 108, connected to the brake cylinder 44 by a pipe 48 and having an escape 109 to atmosphere at its lower end. Two opposed' valves 104, 105, are in the lower chamber on a stem 54. One valve, 104, controls the opening between the two chambers; the other, the opening from the lower chamber to atmosphere. When the stem is upward, the-Valve 104 is closed and the valve 105 is open,, with the result that no compression reaches the brake cylinder 44, and the automatic device is idle. When the stem is down, valve 104 is open and valve 105 closes, with the result that compressed air flows freely from the upper chamber 106 through the lower-chamber 108 and the pipe 48 into the brake-cylinder 44, forcing the piston to the right, with a resultant application of the brakes. In the normal position, the stem is up, excluding compressed air. Thus it is clear that operation of the brake-setting unit is entirely and solely dependent upon the slight downward movement of this EPV stem 54. Air pressure from 106 against the valve 104 and a spring 110 normally tend to force the stem down; but the stem is controlled by the-position of an insulated disc contact 102,. capable of bridging- fixed contacts 103-103, attached to the lower end of the stem and this-position is governed by a magnet 53. When this magnet is energized, it brings up the disc-102 bridging the contacts 103-103 and thus-holding the stem in the normal “up position”;. when this magnet is de-energized, the stem-drops to the braking position. Thus the movement of the stem is entirely and solely dependent upon whether the magnet is energized or de-energized. The normal condition thereof being energized, the brake application! depends upon de-energizing this magnet. This brings us to the electrical arrangement governing the condition of this magnet.

The Circuit.

The magnet is controlled by a single circuit which must be influenced to affect the magnet. All other parts, track, reset, hold-off, and protective, to be effective, must influence that circuit. They are all hooked into the circuit for that one purpose. That circuit is normally closed, thus keeping the magnet energized. The only way to affect it so that the magnet will become de-energized and the brakes applied is by opening that circuit. Therefore the other electrical parts are designed solely to break or to prevent breaking of that circuit; either one or the other is the sole function of the track element, the hold-off, the reset, and each of the protective devices. We will describe this circuit, and thereafter, in order, the devices which may affect it.

The prime element in this circuit is, of course, the coil energized magnet 53 which controls the EPV stem 54. Eliminating enumeration of mere conductors, the circuit with *797its possible breaking points is as follows: From ground • through a disc contact 102 on the valve stem 54 which contacts with two points 108) through battery 131, through magnet 53 coil, through contacts 36-37, in the hold-off mechanism, through contacts 122-123, in the reset mechanism, through front contact 71 off relay Q, to ground. This circuit includes, in series, the contacts 102-103, 36-37, 122-123, and front contact 71. Each of these breaking points in the circuit exists and was so located for a definite purpose. The front contact 71 is the breaking point affected by the track signal warning impulse, which operates to open the circuit at 71. The disc contact 102-103, which breaks when its carrying valve stem drops upon de-energization of the magnet 53, is to prevent re-energizing of the magnet 53 by the closing of 71, which is opened only momentarily by the trank impulse, therefore, .'requiring manual operation of the re-setting device. Contacts 36-37 and 122-123 are to prevent undue holding, tying, or manipulation of the hold-off device of the reset device, respectively; these two contacts are purely protective of the device, and will be described in connection with the hold-off and with the reset, respectively. The front contact 71 is the one through which the warning impulse comes from the track element. That contact was connected, by a series of quick-acting relays, to an engine carried receiving magnet. This magnet was so located on the engine or tender that it would pass several inches above and over track magnets by which it might be influenced. While the track unit, including the track magnets is not involved here, it is necessary to have an outline understanding thereof in order to appreciate its relation-to Howe’s purpose and also its relation to and effect upon the engine unit.

The Track Unit.

The purpose of the track unit was to pick up adverse block signal indications and to transmit them to the engine unit, where they were transmuted into action tending to apply the brakes. As said just above, the track influence came directly to the engine through the influence of a track magnet which affected an engine magnet passing over it. The entire engine unit was based upon normally closed main and relay circuits, the engine receiving magnet being in one of these relays. If the track magnet were energized, it would not affect the relay engine magnet. If the track magnet were de-energized, it would cause a flux in the relay engine magnet which would result in opening the front contact 71. Therefore the track magnet was in a normally closed circuit. Each of these track circuits was arranged in blocks corresponding to the block signal circuits, and was so hooked up with the signal circuits ^that it would respond thereto. The result of such response was to open the track circuit and de-energize the track magnet whenever the corresponding signal circuit took a warning position. The exact electrical arrangement of the signal and the track circuits is not in question in this case and need be no further described.

Having sufficiently described the engine and track elements designed to cause automatic service brake application, we pass to the apparatus to reset the device after such automatic application.

The Reset.

A service application always requires use of the reset unit to release the brakes and to reset the automatic braking apparatus. Here the reset unit U is hooked into the main circuit and operates as a shunt around the disc contact 102-103. This shunt is created by closing the normally open contact 115 — 116, and is made up of ground conductor 141, contact 115-116, and conductor 140. Its operation is and can only be by manually pressing down the plunger 111 until the lug 114 at its lower end presses the spring contact 115 against contact 116. The result is that the shunt thus established closes the main circuit around the then open disc contact 102-103, the magnet 53 re-energizes, and the disc 102 is raised by the magnet armature and held in contact with 103-103, thus closing the main circuit and valve 104 and opening valve 105, permitting compression in cylinder 44 to escape to atmosphere through 109.

Obviously, if this reset plunger were fastened down, it would completely nullify the service braking unit, because it would establish a branch circuit as to the magnet 53 and thus prevent de-energizing thereof by opening one of the branches at 71. To prevent such manipulation, Howe provided self-protection for the reset unit. The unit was within a strong inelosure 112, so that access was denied, and all the engineer could do was to operate the plunger 111, which, if not held down, would be raised by the plunger spring 117. Hooked up with this plunger and entirely within the box was the safety mechanism. Its principle of operation was that, if the plunger were held down closing the contact 115-116 longer than a brief predetermined period, the normally closed contact *798122 — 123, located within the box, of the main circuit, would be opened and held open until the plunger was released and-the spring contact 115 rose away from contact 116. This result was brought about as follows: To a pivoted lever 119 was attached a coil spring 120, which constantly tended to pull the insulated end 121 of the lever so that it would break the contact 122-123. The other end of the lever rested on a lug 118 on the plunger so that the stronger plunger spring 117 would prevent movement of the lever unless that resistance was removed by pressing down the plunger. As soon as the plunger was pressed down, the lever spring 120 became active, and the insulated end 121 began to rise to break the contact 122-123. The speed of this movement was controlled by the timed passage of a piston 126 through the cylinder 124, which passage was retarded by the resulting compression therein, of ail which eould escape only through the restricted orifice 125. The size of the cylinder and of the orifice and the strength of the spring 120 eould be so arranged as to constitute the piston cylinder a timing device. If the plunger were held down longer than the time thus predetermined, the contact 122-123 would be broken and the brakes set. •

Thus the reset operated to release the brake and to reset the automatic braking device after an automatic service application, and was self-protected against manipulation. It must aet in the way intended, and such action eould not be defeated.

The location of the reset was a matter of importance to Howe’s purpose. That purpose was to inflict a “penalty” upon the engineer by giving publicity of his neglect to the fireman and by causing the fireman annoyance; thus inducing the engineer to be careful in observing the signals and operating his hold-off (patent p. 10, lines 78-81). His method of accomplishing this purpose was to continue the automatic application of the brakes “until the fireman, or some person other than the engineer, releases this pressure” .(patent, p. 9, lines 56-58) by operating the reset.6 He calls this “the fireman’s reset device” (patent, p. 9, lines 90, 129, 130, page 10, lines 83, 84). It was therefore located where it eould be reached by the fireman but not by the engineer “while he retained hold of his brake valve handle” (patent p. 10, lines 62-67). Thus the engineer might, without operating the^ hold-off, forcibly hold his brake valve lever against the automatic pressure and thus annul an automatic" application already started until the fireman released him by operating the reset. While the patent does not suggest it, the reset might be so placed as to be inaccessible until the train was stopped, and this could be within the patent, as it affects merely the penalty on the engineer.

Thus far we have described the apparatus and operation by which Howe provided automatic service braking, reset the brake applications, and protected the reset. It remains to describe his means of enabling the engineer to keep complete control of the train and prevent the initiation of automatic braking.

The Hold-Off.

There were two situations which might arise where the engineer eould affeot the automatic braking. If the automatic braking had started, he could forcibly hold his brake lever against the pull of the piston 45. This pull would continue, since the pressure was continuous, until the reset was operated by the fireman. In this ease, holding the ordinary brake lever against the automatic braking would effectually nullify!, not forestall, a braking operation already initiated.7 The other situation was where the engineer observed the block signal, and, in response thereto, prevented any automatic application by timely and proper operation of the-hold-off apparatus. The hold-off apparatus (unit H) functions through nullifying the de-energization of the EPV magnet 53 because of a break in the main circuit by the opening of front contact 71 therein when an adverse track impulse is received. This is accomplished by a shunt away from 71. Normally this shunt is open and ineffectual. When closed, it is effectively operative. If that contact is closed when the contact 71 momentarily opens because of an adverse track impulse, the main circuit will be undisturbed by that impulse. The closing of this open contact in the shunt is the vital function of the hold-off. This may be easily done by the engineer by shifting the lever 18 of the hold-off so that the contact 22 bridges the opening 23 in the shunt. The shunt is made up as follows: Hooks into the main circuit at 135, *799conductor 138, contact 22-23, and conductor 139 to ground.

The remainder of the hold-off apparatus is the mechanism to protect it from misuse. The desirability of such protection is evident because, without such, the engineer could readily fasten the lever 18 in an operative position and thus entirely, continuously, and permanently prevent all automatic brake operation. In addition to strongly inclosing the entire hold-off, except the lever handle, the protective feature is intended to act to automatically apply the brakes whenever the hold-off lever is held in operative position beyond a brief predetermined period. The mechanism consists of an element which will break the main circuit, thus de-energizing EFV magnet 53, when the hold-off lever is held operative beyond the prescribed time.

This protective feature and its operation are as follows: The lever 18 controls a rotary valve V, which is so constructed that, when the lever is moved to operate the hold-off, the valve opens and admits compressed air from pipe 24 into pipe 27, where it passes through an adjustable orifice 28 into a small reservoir 29, from which it flows through pipe 30 into cylinder 31 against piston head 32, which is attached to a piston rod 34 carrying a lug 35 at its outer end. As the air force overcomes the opposition of the compression spring 33, the lug travels toward one element 36 of the contact 36-37 in the main circuit. If the valve V remains open long enough, lug 35 will press spring contact 36 away from 37, thus opening the main circuit, which will de-energize EPV magnet 53 and automatically apply the brakes. The valve V contains a recess through which compressed air which has been admitted may be exhausted to atmosphere, when the lever is in its inoperative position, thus permitting the spring 33 to force the piston 32 into its normally inactive position, thereby carrying the lug 35 away from the contact 36. The result is that the engineer must return the lever in less time than it takes the lug to travel to and break contact 36-37.

Summary of Howe.

From the above description of the various parts of the engine element of Howe, now in suit, it appears that inaction, however induced, of the engineer at an adverse block signal, will result in an automatic brake application which cannot be released by the engineer while he controls his usual brake valve lever, and will eventually stop the train unless the fireman acts to prevent it doing so; that, once initiated, this automatic braking application ean be stopped only by operation of the reset apparatus; that this application may be entirely prevented only if the engineer, within a definitely limited time shortly before passing the block signal, operates the hold-off apparatus; that the effective operation of the hold-off depends upon observance of the block signal; that the entire engine device, here involved, is so protected that it cannot be prevented from acting in the manner contemplated; that it contemplates that the engineer may forestall automatic brake application or, in conjunction with the fireman, terminate such application.

These results are accomplished through automatic operation of the regular engine brake valve by a single braking unit controlled by a single electro-pneumatic valve operated by a single closed circuit which may and must be affected; that the operation, proper or improper, of all units (track impulse, the reset, and the hold-off), is upon this circuit; that improper operation of either reset or hold-off results in its own defeat.

Claims Involved.

Howe contends that his device is protected by ten typical claims which are infringed. Those are claims 9, 10, 23, 30, 34, 43, 48, 56, 57, and 58. Counsel for Howe state:

“For convenience these claims may be divided into groups and one claim taken as typical of the group, as follows:

“Claim 9, typical of 9, 34 and 10.

“Claim 30, typical of 30, and 43.

“Claim 58, typical of 23,48, 56,57 and 58. “Of the elaims relied upon, claims 9 and 34 are the broadest.”

Those three claims (9, 30, and 58), selected as typical, are as follows:

“9. In an automatic train control system the combination with car-carried brake control apparatus co-operating with the air-brakes of the car to produce an automatic brake application, of means for automatieally initiating the operation of said apparatus from the trackway under unfavorable traffic conditions; a hold-off device conveniently accessible for operation by the engineer and effective only if operated prior to the initiation of operation of said apparatus, for preventing such operation in spite of its control from the trackway, said hold-off device if -kept in the hold-off condition longer than a predetermined limited time after each operation thereof acting automatically to actuate said apparatus and cause a brake ap*800plication, whereby the engineer may prevent an automatic brake application only if he anticipates the operation of the brake control apparatus by a manual operation of the hold-off device and restores the hold-off device to normal within a limited time.”

“30. Railway traffic controlling apparatus comprising an electro-responsive device on a vehicle, a vehicle carried' circuit for said device including a source of current and a normally closed contact, devices located at intervals in the trackway for opening said contact], Imanually operable means on thle vehicle for closing a branch around said contact and effective for a given interval of time to prevent de-energization of said electro-responsive device by a trackway device, and means controlled by said electric responsive device when de-energized for causing a brake application.”

“58. In an automatic train control system, an electro-responsive device on the train controlling the air-brakes and having stick contacts opened upon operation of the device, a stick circuit for said device including said contacts, automatic means for intermittently and temporarily making a break in said circuit during the movement of the train along the track when dangerous traffic conditions exist ahead, a manually operable con-tactor on the train for closing a shunt around said break and thereby maintain said device energized independently of the trackway control, said contactor being incapable of restoring said electro-responsive device after it has once operated and if held in its operated position longer than a limited time independently interrupting said stick circuit, and other means on the train operable to restore the electro-responsive device after operation thereof.”

Having outlined the form of Howe’s device which he alleges is infringed and having set forth the claims upon which he relies, we pass to the citations charged by defendant to be anticipations.

Prior Art Citations.

Defendant urges ten citations as having been suggestive in the prior art. One of these, Sprague, British 18,213, supplemented in some respects by Sprague, 660,065, Spra-gue, 716,953, and the Jones installation, is urged as a complete anticipation of Howe’s entire device. Two others, Barberie, 825,289, and Lydall, British 1,517, are urged as anticipating much contained in Howe, while Sprague, 660,065, Sprague, 716,953, Price, 923,297, Oler, 1,116,320, Pomia, 1,177,941 and 1,177,942, Bulla, 1,387,233, and the Jones and Stephens installations are urged as suggesting certain different features or elements found in Howe.

It is very evident that there are numerous devices providing for automatic train control not referred to in this record (see 69 I. C. C. 258; 85 I. C. C. 403;. 911. C. C. 426), but we should and will confine ourselves to examination of such as are a part of this record or such as were introduced at the trial and used in oral argument without objection.8 In examining the citations, we will first consider the citations (Price, Oler, Pomia, Bulla, and Jones) aimed at particular elements or features of Howe; next, the citations . (Barberie and Lydall) deemed by counsel as, anticipating much of Howe; and, finally, the eitation (Sprague,' British) urged as anticipating Howe entirely —with the last, such discussion as necessary of the two other Sprague patents, which are in other arts.

Price No. 923,297, June 1, 1909, covered “Automatic Train Controlling and Signaling Mechanism for Block Signal Systems.” This patent disclosed a device for automatically blowing a whistle and for setting the brakes through operation of the engineer’s brake valve. Also, it has a device to nullify the action of the brake setting after it has originated. The meehanism for accomplishing these results are similar to Howe, in that an electro magnetic impulse is received by the engine and transmuted into action upon the air brake system. But this is a necessary line of action in all inductive systems. The annulling device is entirely different in theory of operation, natural force employed (being pneumatic instead of eleetrie), and mechanical device.

Oler, No. 1,116,320, November 3, 1914, covered “Railway Traffic Controlling Apparatus.” It disclosed the general idea of a venting of air from the air line through a valve controlled by an energized magnet-which would become de-energized upon passing over a track unit if danger were present. It has no forestalling or annulling device, and is reset by manual restoration of the electric circuit which energizes the magnet valve. It is a forerunner of one feature of Howe, but is designed as a device beyond all eon-*801trol of the engineer/ except this manual resetting.

Fomia, No. 1,177,941, April 4, 1916, is for a “Railway Traffic Controlling Device.” The revelation is of a purely mechanical device which makes use of the air in the brake system for its operations. It is suggestive of several thoughts and purposes worked out by Howe, but the method of accomplishing them is different. For example, it requires alertness of the engineer to prevent operation of the brakes, but this is required at every signal post, irrespective of the presence of danger. This theory is not at all that of Howe, who required, instead of constant attention, action only in the presence of indicated danger, A forestalling device was revealed which must be operated at every signal, and there was the necessary provision to prevent this forestalling being continuous, which would have defeated the whole purpose of the entire device, but such forestalling device and the protection thereof is different from Howe in principle and method, and like Howe only in that the brakes are set if the forestalling device is operated beyond a predetermined time. All that Howe could have gotten from this patent would have been the idea of having some'method whereby the brakes would set if the forestalling device were operated an undue time.

Fomia, No. 1,177,942, April 4, 1916, differs from No. 1,177,941 in that, instead of a contact operation, there is a magnet on the engine which is directly operated by a track magnet when danger is present as indicated by the block system. There is no electric circuit on the engine.

Bulla, No. 1,387,233, August 9,1921, covered “Safety Attachment for Automatic Train-Stops.” The principle revealed is to vent the train air line through a pipe with a valve which is opened when a magnet is de-energized; permitting compressed air from another pipe to flow through a valve, closed by the energized magnet, into a cylinder which operates a piston connected with and opening the valve first mentioned above. To forestall such action, a lever might be thrown which would break the circuit to the ground and prevent de-energizing by the track impulse. To prevent abuse, this lever was attached to a normally closed valve in the pipe leading to the train air line in such manner that the valve would be opened and the train line “bleed.” This differs from Howe in method (direct venting of air from train line to set brakes instead of through regular vent operated by engineer’s lever), in the device to carry out, such idea, somewhat in the operation and device to prevent de-energizing of the magnet, and in the method and device to ^protect the hold-off from abuse (Bulla bleeding the train line through a valve while Howe de-energizes the EFV magnet with the same result as though there had been no holding-off attempted).

■ The Jones citation is of a train control device first presented in 1907, and tested by the Division of Safety of the Interstate Commerce Commission in 1911 and 1914. The evidence in this record appears in the form of a report by the chief of the above division; an extract from the Signal Engineer regarding such report and an extract from the Railway Signal Engineer, about four years later. This device employed an energized magnet controlling a valve affecting the air line. This magnet was de-ener-gized when a controlling circuit was meehan--ieally broken when a trip on the train shoe struck a de-energized contact rail. There was a reset handle by means of which the valve magnet armature might be manually restored to normal. To prevent nullification of the entire device by this handle being fastened in the reset position, the handle was attached to an air brake cock opening a train line exhaust port, which would remain open until the handle was returned to its normal position. This device had the principle and method of an air valve controlled by. an energized magnet, but the controlling electrical arrangement is very different and more complicated than in Howe. There was no hold-off device in any true sense, and the reset device, even if unprotected, would not have been such, but would have been capable of annulling the entire automatic braking device. The reset was purposed solely to reset the braking device, and was not at all intended to be used as a hold-off, although it is conceivable that the engineer might, after the operation of braking started, annul it by a prompt reset. The protecting element of the reset was solely purposed to prevent complete and continuous annulment of the entire braking device, and the method thereof was entirely different from Howe; Jones causing a direct venting of air, while Howe broke the electric circuit which energized and kept in place the valve magnet.

• Barberie No. 825,289, July 10, 1906, was a “Safety Apparatus” to stop an electric car or a train passing an adverse block signal or semaphore. It was made up of two combined, and eoaeting unite — one to set the brakes; the other to shut off the power. The *802brake-setting unit wás entirely, pnuematic. The power suspension unit was eleetropneu-matie. The outline idea was as follows: An adverse signal would mechanically raise a track trip so that it would strike and break a destructible cap projecting downward from the electric ear or the locomotive; the breaking of this cap caused a movement of compressed air in an arrangement of pipes and cylinders which eventuated in an exhaust of the air brake line and a consequent setting of the brakes; the same movement of compressed air moved a valve contact breaking an eleetrie circuit and resulting in shutting off the power. There was. a forestalling and also a resetting device so combined as to be operable by a like movement of the same hand lever. The forestalling device was pneumatic with an oil cylinder timing element. It operated to draw the destructible cap out of position for contact with the track trip, and, through working of the oil timer, would lower it again into position for contact. The reset device operated pneumatically by establishing the normal stream and condition of compressed air which would prevent, escape. This re-establishment of normal moved the contact valve back into normal and released the power. If the motorman should fail to return the lever to the normal operative, position within a few seconds after forestalling a trip', an air cylinder operated to open the air line, and, as a consequence thereof, shut off the power. The destructible cap was to be replaced after each breakage. Barberie reveals the ideas of immediate forestalling just before passing the danger signal and of time limiting that forestalling with a penalty of application of brakes for violation thereof, but bis method and mechanism for working out such ideas are very different from Howe. His entire braking method is different and in no way suggestive of Howe.

Lydall, British No. 1517 of 1908, is a safety device with reset and protected annulling element. It is a contact type wherein every track element mechanically breaks an electric circuit which results in release of a solenoid which falls, striking the end of a lever, to the other end of which is connected a valve opening into the brake air line. The solenoid is controlled by two interlocking circuits, both of which must be broken. The hold-off results in closing one of these so as to shunt the current around the break caused by the track contact. This shunt is accomplished by a lever which closes a contact. If the lever is held more than a few seconds there is a connecting device which starts a whistle which continues to blow until the lever is returned to normal. The electrical arrangement to operate the brakes is not the same as Howe but has points of similarity. The hold-off has the'same purposes as Howe, and operates through a shunt circuit, but is differently arranged, and has a different and differently arranged penalty for misuse. The reset is by manual replacement of an armature, and is essentially different from Howe.

The above citations have been only briefly discussed in this opinion for several reasons. Any train control device is necessarily rather complicated, and to fully describe and compare with Howe in detail each of these citations would unduly extend this opinion without serving any useful purpose, since, with the exception of Barberie and Lydall, the above citations are intended only as to separate features of Howe. We have thought it advisable to make careful analysis of each of such citations and a careful comparison with Howe and then to state only a few of the distinguishing features. We have done the same as to Barberie and as to Lydall, and, while the citation as to each of those two is intended to take a wider scope, we have deemed it sufficient to state only our conclusions as to each.

The Sprague patents, or at least No. 18,-213, are so emphatically the reliance of ap-pellee that they should be more particularly stated. The main emphasis is put on Spra-gue, British No. 18,213 of 1915, the other Sprague patents being outside of train control devices and relied upon merely as anticipating certain features of Howe.

Concerning Sprague, No. 18,213, counsel for appellee says: “The patent discloses substantially everything that Sprague is doing and that Howe is claiming.” And again; “The claims in suit in so far as they depart from the disclosures of the Sprague anticipation are directed to details of engineering development which do not rise to the dignity of invention.” The Sprague patent is described by counsel for appellee as follows: “To sum up the disclosure of the British patent, it is, as already stated, an automatic, inductive, intermittent system of auxiliary train control.'. Impulses received from track magnets will cause brake applications of two different degrees or types. One type may be forestalled by timely action on the part of the engineman, and the other cannot be forestalled. There are also > provided automatic resetting means which are operable if the brakes have been applied automatically as the result of a single track impulse. There are also manual resetting means inaccessible *803to the engineman which are required if the brakes have been, applied by a double impulse from the track.”

Sprague intended both a train-stopping and a speed-control (without stopping) device. We are not concerned with the purely speed-control features and “will confine ourselves to the train-stop portions as revealed in the patent 18,213. That plan was as follows : Track magnets so spaced and so electrically charged as to operate, through induction, upon a primary and upon a secondary feature of the electropneumatic braking element on'the engine. These track magnets operate in pairs, possibly in series also. Upon passing the first or “distant” (from danger) influencing tracks magnet, visible and audible warnings are given in the cab and these warnings are very shortly followed by a “primary” or partial service brake application which moves the engineer’s manual brake control handle to the “lap” position, where it remains until released by the engineer. At the same time, there starts a device whieh records the clock time and the character of the automatic application. Upon reaching the next (“home” or near danger) track magnet, if it is then in an electrically influencing condition, a second brake application may take place. Whether this second application shall take place, its character and results depend upon what the engineer does or fails to do between the time of the warning and the time the second -magnet is reached. These are set forth in the patent on pages 6-8 and copied in the footnote,9 A summary of the quotation in the footnote as to the action of the Sprague device is as follows: The timed commutator *804governs the result of the secondary track influence; if the speed is not slackened, automatically or manually, so that the commutator runs its full course before the second track impulse is reached, that impulse will stop the train and cannot be prevented; if either the automatic primary application or the manual braking by the engineer, or presumably if from any cause the then speed of the train, results in not reaching the secondary impulse until the commutator has completed its full course, then only the limited service application will result from the second track impulse; the engineer may prevent the automatic braking from the first impulse, but only by application of the brakes within the brief time between the cab signals and the primary brake application; the engineer may nullify the effect of the primary application, after it has begun and before the second impulse is received from the “home” magnet, but only by forcibly bolding the brake valve handle against the pressure of the primary application; the only way the engineer can avoid the braking action from tbe second impulse is to manually reduce tbe speed between tbe two impulses so that tbe second impulse is reached after the complete commutator cycle.10 Iu short, where the two track impulses exist, the device will apply the brakes unless the engineer, within tbe very brief time limit, takes active control and himself applies the brakes so as to bring the speed within the required limits of the device. The commutator sets the limit, and the device will enforce that limit, and the engineer can only act to comply with a maximum of speed not exceeding that limit unless he wishes to be stopped at the second impulse. There is no way in which the engineer can, in accord with the theory and purpose of the device, entirely escape the braking action— automatically or manually. If he desires to prevent the primary impulse, he must do so by applying the brakes. If he annuls that impulse after it has started, such can be only for the short distance before reaching the secondary impulse, which will stop the train. The device purposes and all action contemplated by it will compel some application of brakes, automatically or by the engineer, if both magnets are impulse giving.

The differences between the theory, the practice, and the results of this system and that of Howe are distinct. The Sprague theory is to compel the braking operation either by the device or by the engineer whenever a danger situation is indicated. The engineer may avoid stoppage of his train, but he cannot avoid slowing down to a maximum prescribed speed. The penalty for lack of alertness is the written record which reveal to his superiors just what occurred and when and which may be used against him. The stoppage, if it happen, carries no. additional penalty, because all he hás to do is manually close the secondary exhaust valve which is not required to he placed where he must get down off the engine or otherwise greatly inconvenience himself.11

*805In short, he may act with the device or he may let it control, with the sole penalty of having his inaction made matter of record. Also, if he attempts to act against the device and there be an impulse at the second magnet, his attempt to frustrate is recorded. Another distinguishing feature is that the Spra-gue device is in no wise dependent upon a block signal system, but may be operated where there is one or equally well where it is not. It is a complete device within itself. Even if it be employed with a block signal system, the engineer may disregard the block signals entirely and depend solely upon the cab signal warnings provided by Sprague.

The Howe plan was for the device to take the train entirely away from the engineer unless he forestalled its operation. Such forestalling depended entirely upon the observation of block signals by the engineer. It applied to each successive separate signal and required a separate forestalling for each. The result of forestalling was entire avoidance of any effect from the particular track impulse forestalled, thus leaving the train in ás .complete control of the engineer as though the track impulse were not present (he could increase, slacken, or continue the then speed). Instead of resisting an automatic brake action, he entirely prevented such. In fact, he alone had no power to affect a braking action once automatically begun.

The purpose of Sprague was “to duplicate as- far as is practical the action of the engineer in making service application, while at the same time leaving him free to make further service or emergency applications, or under certain conditions to release his brakes, or to anticipate possible automatic braking by proper manual braking.” Patent, p. 16, lines 55, 56, and page 17, lines 1, 2. The purpose of Howe was to encourage and compel the engineer to be alert to observe and suitably act upon the indications of block signals. The penalties were entirely different, being a permanent reeord of inefficiency in Sprague and a temporary annoyance and inconvenience in Howe.

In working out their purposes, both employ some elements which are similar and which were, separately, old in electric and mechanical art and in this particular art before either entered it. But these quite different purposes necessarily resulted in distinctly different arrangements and in the use of features not common, so that the two combinations are essentially different.

The citations of Sprague, Ho. 716,953, December 30, 1902, for an electric elevator, and Sprague, Ho. 660,065, October 16, 1900, for a traction system, are intended to anticipate the Howe feature of a fixed time limit beyond which the forestalling lever may not be left in that position — that is, a time limit protection of the forestalling device from misuse. The idea may be in these patents, but it is not an idea, but the means of making an idea of practical use that is patentable. The method of both is different from Howe.

The citations are not anticipations of Howe. Some of them contain one or more features of construction like elements in Howe, and others have some of the same ideas which Howe worked out in a different way, but no one nor all of them can anticipate the Howe device. We find the Howe patent valid.

Scope of Howe Patent.

The range of equivalents to be allowed a patent is quite a different matter.from its validity. Inventors naturally state their claims in as broad and general language as can secure allowance of a patent. Construction of such broadly stated elaims, necessitated in infringement cases, often places practical bounds upon such general language. This is particularly true of combination patents which from their very nature, usually include or may be entirely made up of elements old in the art. Were it advisable, there is hardly any way in which such character of claims can be definitely defined in a single infringement action so that such definition may stand as all-inclusive of other and different devices than those involved in the immediate suit. Generalizations are safer and of more real use.

Howe entered a field which had been and was being industriously cultivated by others.. The safety of railway passengers, freight, and equipment had long been an apparent necessity. The desirability of a device which could detect danger to a moving train and avert disaster by stopping the train was very evident. The citations here made and the reports and action of the Interstate Commerce Commission reveal the wide-spread serious efforts of the Commission, of the railways, and of numerous inventors to meet this need. Before Howe entered the field, much had been done to determine what the problems of train control were and much had been done in endeavors to solve those problems in a practical way. While there was one common purpose of safety, the ideas as to *806what would give protection were different, as well as the methods of making a given idea of safety practical and workable through some device. Howe’s idea of protection was to leave the train completely in control of the engineer if the engineer was alert and attentive and comprehensive of the danger situation, but to take control from him if he was not. His device was intended to secure those results, and apparently does. But this idea was not new, and others (see discussion of citations above) had aimed at and seemingly reached either or both of the above results. Also, in train control devices, whether striving for the same results as Howe or not, there had been various solutions of the problems of train control, which problems were automatic stoppage, hold-off, or annulment thereof, reset of the automatic device after a braking operation, and protection of the various function elements from interference or defeat by the engineer. Also the range for invention was somewhat confined by the ever-present situation of transmitting some warning from the track to vent the air brake line and hooking into that the hold-off or annulment and the reset with the protective features as to all. Also electromagnetic action, induced from the track, on valves controlling the air line, had been described, as well as some other features. All of these considerations placed limitations upon what Howe could appropriate.

On the other hand, his plan and device Were distinctly different. His was the first to center entirely on the action of one electropneumatic valve, influencing the brake air line, and one main electric circuit controlling that valve where the brake setting, the reset, the hold-off, and the protective features all centered on that one circuit to actuate or prevent action of that valve: In such a plan he was first. While a plan — a conception — is not patentable, yet where that plan is in the arrangement of mechanical parts and of forces, that arrangement is entitled to a fairly liberal range of equivalents, where the inventor was the first to conceive the plan and the first to effectually make it workable.

• Infringement.

Determination of infringement naturally depends upon a comparison of the two devices. This comparison is to be made having in mind that Howe entered a crowded field, but that he first evolved a plan of centering all of his device upon a single eleetropneu-matio valve, influencing the brake air line, and one main electric circuit controlling that valve, where the brake setting, reset, hold-off, and protective features all centered on that one circuit to actuate or prevent action of that valve. If such is the arrangement of defendant’s device, there is infringement, otherwise there is not.

Defendant’s device is as follows: The brakes are applied by an eleetropneumatic valve. This valve is normally held closed to atmosphere by an energized magnet. Normally this magnet is kept energized through a normally closed electric circuit which includes two separately located contacts — one where the reset is located, and one where the hold-off is placed. The same battery serves a secondary normally closed circuit which includes a front contact, which is influenced by the track-received impulses, and a stick relay; the stick being mechanically connected with the reset contact in the valve circuit. The effect of a track-received impulse is to open the front contact, breaking the secondary circuit and de-energizing the relay coil, thereby causing the stick to drop, carrying with it the mechanically attached reset contact member, and thus breaking the main circuit which de-energizes the valve magnet causing the air valve to open and a consequent brake operation. The reset is by pressing a button which closes the reset contact and its mechanically connected relay stick contact, thus closing both main and secondary circuits, with a consequent re-energizing of the valve magnet and the relay. The hold-off consists of two co-operating features, both set in operation by the movement of a lever— the first bridges a normally open contact establishing a shunt around the front contact which receives the track impulse, thus avoiding the effect of such impulse; the second is a timed movement of a bridge which connects a contact in the main circuit, as the movement of this bridge continues as long as the lever is in hold-off, the contact will be broken if the lever is so held beyond the limit of the timing mechanism, thus opening the main circuit and'operating the brakes. While there are minor differences, such as lack of protection of the reset and in the timing methods of the hold-off, Howe being by air and appellee by mechanism, yet the only difference which attracts serious attention is the use of a secondary circuit. Is such use outside the equivalents allowable to Howe? The traek impulse and the resultant opening of the front contact are momentary. It is essential in both devices to maintain this momentary condition by having it break some other contact which will remain broken and *807thus keep the circuit open. Howe accomplishes this within his one circuit by the disc contact attached to the electropneumatie valve stem which, by weight and gravity, will remain open until the reset has closed the circuit and energized the valve magnet sufficiently to draw the disc up into contact. Ap-pellee does it by a relay stick located away from the valve, but mechanically attached to and governing a contact in the main circuit. The disc and the stick perform exactly the same functions and for precisely the same purpose. The only use for a second circuit is, after the momentary impulse has passed, to keep the break until reset. It is as though appellee, had taken the disc contact away from Howe’s valve and placed it in another part of the electric hook-up, with only such additional equipment and arrangement as was necessary to make it perform the same function in its new location. This relocation is such as any electrical engineer would readily conceive, and is not an.inventive difference from Howe, but is within the equivalents to which his patent is entitled. Thus we have, in appellee’s device, the equivalent of one normally closed circuit controlling a brake-operating electropneumatie valve causing brake operation with a reset hooked into that circuit and a hold-off hooked therein and so protected that an undue holding off will operate the electropneumatie valve and work the brakes. With omission of Howe’s protection of the reset, this is the Howe plan and device. It is true that the air element has differences from Howe, but these are more by way of additions rather than of changes.

We think the Howe patent is infringed by the device of appellee, and that the decree should be reversed, with instructions to set the decree aside, to enter a decree of infringement, and to take such other and further action as is in support thereof. It is so ordered.

From January 1, 19X1, to March 31, 1924, the Interstate Commerce Commission investigated 107 collisions, resulting in 501 deaths and 2,417 persons injured, due directly or indirectly to failure of engineers to observe or to he governed by block signal indications. In re Automatic Train-Control Devices, 911. C. C. 433.

In 1922, when the Interstate Commerce Commission first ordered installation of train control devices, it expressly denied the use of such control,saying (Automatic Train Control Devices, 69 I. C. C. 258, 275):

“We have eliminated the provision in the specifications of the joint committee under which the engineman would he permitted, if alert, to forestall the automatic-brake application and proceed. Some of tbe respondents object to the elimination of this provision. They contend that, in many instances, it is proper for a train to pass an automatic block signal in the stop position. Where a train, for example, is being admitted to a sidetrack by a switch tender or tbe crew of another train occupying the main track, tbe signal will be in the stop position and yet it may be proper, it is stated, for the train taking the siding to pass this signal, on authority of a hand signal, without stopping. It is contended that there are so many conditions similar to the one cited that the elimination of the manual-control provision practically eliminates the simple automatic stop from consideration.

“The essential safety function of an automatic stop Is to compel obedience to a stop signal. Under tbe provisions which we have adopted tbe apparatus may be restored to normal condition manually after the train has been stopped and then the train is permitted to proceed. This a necessary safeguard. Where the •device is made subject to the manual control of the engineman so that he may forestall, i. e., prevent, the automatic-brake application according to bis own judgment of the conditions, the automatic safety feature of the device is to that extent nullified. It is assumed, by tbe proponents of manual control, that no engineman, if alert to a dangerous situation, will deliberately cut out the automatic-stop device. The proper use of the manual control would depend, therefore upon the judgment of the engine-man. His judgment would be the determining factor in situations of known or unknown danger. This factor of human judgment is the factor which an automatic train-stop device is designed to eliminate.

The manual-control feature is, in our opinion, a dangerous one which will permit the judgment of the engineman to intervene, and thus may prevent the essential function of the train-stop device, namely, its automatic operation in cases of emergency.”

However, automatic train control devices with this permissive element were installed by the Chicago & Eastern Illinois in 1914 and by the Chesapeake and Ohio in 1917 (91 I. C. C. 426, 436), and the Rock Island was permitted to install such in 1923 (85 I. C. C. 403, 405).

In 1924, the Commission held another hearing and entered an alternative order allowing such permissive element; that order and the reasoning therefor being as follows (91 I. C. C. 426, 435):

“In Automatic Train-Control Devices, supra [69 I. C. C. 258, 264] we said:

“The essential safety function of any automatic train-stop device is to stop a train where a dangerous condition exists ahead of the train, when the engineman for any cause fails to take proper action to stop.

“In that report we eliminated from our specifications the provision under which with an automatic train-stop device the engineman, if alert, would be permitted to forestall the application of the brakes by means of the automatic stop and proceed uhder limits fixed by train order or prescribed by the operating rules of the company or in accordance with hand signals. At the original hearing certain of the respondents objected to the elimination of this feature. The committee representing all the respondents now ask that it be restored for tbe reasons, as-they contend, that without it they are compelled for operating reasons to use some form of speed control. They further contend that the introduction of this permissive feature would eliminate many of the objections that the operating officers now make to the so-called inflexibility of automatic train-stop* devices. It is not necessary in all cases, and at times it is even unsafe, the carriers contend, to stop* long, -freig-bt trains by means of the automatic application of tbe brakes at a stop signal, and in this-contention they are supported by representatives of the employees. When proceeding up a long grade-such stops are often undesirable because of the difficulty of starting tbe train again with its heavy tonnage. When on a down grade, an automatic application of the brakes following an application by the-engineer may waste air pressure to an extent that will endanger tbe safety of the train. Under certain, slack conditions it may result in buckling the train, thereby threatening serious accidents. To facilitate-movement of traffic when approaching junctions, yards, or other similar points, it is often necessary to pass signals at danger under the guidance of hand signals. In such cases the carriers state that a permissive feature should be included in an automatic-train-stop device to permit the engineman, if be is alert and so indicates by performing an affirmative action, such as pressing a button or throwing a. switch in the cab, to pass tbe signal without being stopped, controlling his train by the ordinary methods of applying the brake.

“Tbe matter of providing for the permissive feature in automatic train-stop devices was considered, in our original report. While there was testimony in that case both in favor of and against the permissive feature it was inconclusive. At the hearing, in this case the testimony was overwhelmingly in favor of the permissive feature. Operating men almost without exception favored the adoption of such. a feature and expressd the opinion that it was sufficient to require the engineman to take some affirmative action to indicate that he is alert, has knowledge of the signal indication, and is operating his* train in accordance with the operating rules. The committee representing the carriers were a unit in favor of the permissive feature. The chief operating officer of the Rock Island and one of the locomotive engineers from that road who appeared as witnesses for the train-control companies favored^

*793■the use of the permissive feature, and some of the representatives of the train-control companies stated that it was a desirable addition to a train stop. 'Other representatives expressed a contrary view. Certain carrier officials récognize the possibility ■that this feature might lead to carelessness, but believe that it should be left to the judgment of the -management of a road to decide whether a permissive feature should be employed under certain op•erating conditions.

“The installation on the Chicago & Eastern Illinois, which has been in operation on a full division since 1914, and the installation on the Chesapeake ■& Ohio, which has been in operation since 1917, have the permissive feature and no instance has developed •where safety has been adversely affected thereby. Both of these companies favor its continued use ■for reasons above stated, which they have set forth -at the hearings and in petitions which they have filed for a modification of our first order in this respect.

“We are of the opinion that the evidence now before us warrants a modification of our former conclusion with respect to this permissive feature, although we shall continue to keep this matter under close observation. Paragraph 1, under the subhead ‘Functions' of our first order, will therefore be modified to read as follows: 1. Automatic train stop:

“(a) Without manual control by the engineman, requiring the train to be stopped; after which the ' .apparatus may be restored to normal condition manually and the train permitted to proceed; or—

“(b) Under control of the engineman, who may, if alert, forestall the application of the brakes by •the automatic train-stop device and control his train in the usual manner in accordance with hand ■signals or under limits fixed by train order or prescribed by the operating rules of the company.

“The above modification requires no departure from the specifications and requirements contained in our first order; it merely provides an alternative feature which may be adopted if desired.'*

Some of the various steps may be found in the three opinions of the Commission (69 I. C. C. 258; 85 I. C. C. 403 ; 91 I. C. C. 426); others -will appear in the discussion of the citations against the patent in suit.

That form provided for an automatic service application at the caution signal, and between that and the stop signal an emergency application if the speed had not been reduced to a predetermined rate between track elements suitably placed between the two signals. The engineer, through a hold-off device might prevent any service brake action at the caution signal, or he might annul such action after it had begun by holding his brake lever against the pull of the automatic device for a predetermined period, after which period the pressure ceased. He had no control over the automatic emergency braking except by reducing his speed» to the maximum permitted between the two track elements, intermediate of the caution and danger signals. The penalties provided in this form, if he failed to forestall, were, at the caution signal impulse, holding, for a limited period, his brake lever against the pull of the device or permitting the train to be stopped by such application —to this might be added a cab whistle which would notify the fireman of his failure. If he forestalled or annulled at the caution signal, but failed to slacken speed to the required maximum a's he drew toward the danger signal, that failure was penalized by an emergency or a heavy braking stop with a» slight delay in release of brakes. The same held-off unit was used in each of the two forms. No reset unit was included in this form, because both the service and the emergency applications were followed by an automatic release of brakes and reset within a time-controlled period. No emergency brake unit was included in the other form here involved.

References are to Insert which is Fig. S of the patent.

Also the patent'(p. 2, lines 8-14) reads: “Fig. 5 illustrates a modified construction of the car equipment in which the cooperation of. the fireman, or some person other than the engineer, is required to get the engineer out of the difficult position in which he is placed when he fails to make timely operation of his hold-off device.”

In the form of automatic brake cylinder and EPV arrangement sh.own in Fig. 2 of the patent, this cylinder pressure would disappear in a- predetermined time. Not that form, but the one shown in Fig. 5 is in siiit.

Sprague, 660,065 and 716,953, were introduced at tlie trial, but not included in the transcript to this court. At oral argument, a blueprint of Figs. 12 and 13 of 660,065 and printed Patent Office copy of 716,953 were used, without objection, and will be regarded as before us.

“The practical operation of the above outlined equipment can be explained-in a general way by considering its use in connection with an automatic signal system for the purpose of protecting automatically signaled block sections, in which the braking of the train may be conveniently considered as divided into three periods.

“First: When the control is entirely in the hands of the engineer, and full responsibility for train handling, both observation of signals and application of brakes, remains in his hands without interference of any kind by automatic control;

“Second: When the responsibility of signal observation and train control is divided between the engineer and automatic apparatus ready to make a regular service application, during which period, and late enough not to remove the sense of responsibility from the engineer, the automatic will take full charge if the engineer fails to do so, but If not will permit him to maintain full control; and “Third: When the braking is increased in effort, and release control of brakes, if the signal is still against the train and it has approached or is approaching the advance block at too great a speed, is taken out of the engineer’s hands until the train has been brought to- a¡ stop.

“For illustration, we will consider the second and third periods, and assume an advance block occupied, with the proper condition in the track circuit of the automatic signal system, but for some

reason the engineer does not make a manual application of the train brakes.

“In such case, as the locomotive approaches and passes at a speed above any predetermined maximum speed over the first or distant track magnet (located at a proper distance from the advance block), the latter having become energized on its approach, a magnetic flux will be transmitted through the inductive receiver on the locomotive. This results in establishing a momentary localized electromotive force in the receiver, creating a current which is transmitted through the coil of the electric detector relay, which immediately indirectly initiates the revolution of the commutator in the cycle train, changes the electric cab signal from clear to danger and starts the cab audible alarm. Should the engineer take immediate advantage of this cab signaling to manually apply the brakes, then no automatic brake pipe reduction or recording of the same will obtain. However, should he neglect the warnings aind not make a prompt manual application of the brakes, then the automatic service brake valve will make a predetermined (established by positive adjustment) brake pipe primary service reduction of a definite amount in some definite time, according to the character of train and service, and will set the brakes on the entire train with entire safety and in accordance with established manual practice.

“An adjustment for ten pounds brake pipe reduction will he assumed in illustrating various operations. At the moment of the automatic service brake valve primary application, the mechanism of the engineer’s automatic valve head moves the engineer’s rotary valve to lap position to avoid other than manual releasing of the brakes, — that is, to blanket the main reservoir and feed valve ports. This condition of these ports is maintained by pneumatic pressure, and the automatic application made of the brakes can only be released — prior to the conclusion of the operating time cycle of the cycle train commutator — -by a forcible manual pushing of the engineer’s brake valve handle to release position and the holding of it there or in the running position.

“Also, at the moment of the automatic service brake value operation a distinctive record of the same is made by the automatic brake application recording device, thereby indicating the clock time as well as the character of such automatic application.

“When the train, now retarded by the automatic primary 10 pounds service application, and subject to further manual application, arrives at the second or home track magnet at above a predetermined maximum speed, the brakes are subject to an increased automatic application, dependent upon the conditions then existing, as follows: •

“(a) Should neither manual releasing or further application of the brakes occur prior to reaching the home track magnet and prior to the completion of the operative cycle of the commutator, then an automatic secondary full service application of the train brakes will obtain at the moment of passing the home track magnet. This action will cause the electric secondary exhaust valve to remain open until manually closed, and for that period take the releasing of the train brakes automatically out of manual control. At the instant of this second brake application a distinctive record is made thereof by the automatic brake recording device;

“(b) Should neither manual releasing nor further manual application of the brakes occur prior to reaching the home track magnet, but after the completion of the operative cycle, then only the existing automatic 10 pounds service reduction of the traiin brake will exist at the home track magnet, and its continuance will be recorded;

“(c) Should the given automatic service application of the brakes be manually released prior to reaching the home track magnet and prior to the

*804completion of tbe operative cycle, then an automatic full emergency application of tbe train brake will occur at tbe borne track magnet and be distinctively recorded;

“(d) Should tbe given automatic service application of tbe brakes be supplemented by a manual further application prior to reaching tbe home.track magnet and prior to tbe completion of tbe operative cycle, then an automatic full service application will occur at tbe borne track magnet and a record similar to the last mentioned will be made;

“(e) Should tbe given automatic service application of tbe brakes be manually released prior to reaching tbe home track magnet, but after tbe completion of tbe operative cycle, then a second automatic 10 pounds service application will occur at tbe borne track magnet and be automatically recorded ;

“(f) Should tbe given automatic service application of tbe brakes be supplemented by a manual brake application prior to reaching tbe home track magnet, but after tbe completion of tbe operative cycle, then tbe resulting brake pipe pressure will, subject to any further manual reduction, continue until released, and there will be another service record made at tbe home track magnet;

“(g) Should tbe engineer take advantage of tbe cab signaling, within the lapse of time allowed pri- or to tbe automatic primary service application, to manually apply tbe brakes, then a full service application will occur and be recorded at tbe home track magnet if and when tbe latter is reached prior to tbe completion of tbe operative cycle, but if after such manual brake application, tbe engineer releases or goes to running position, a 10 pound service application will occur and be recorded at tbe home track magnet if and when tbe latter is reached after tbe completion of tbe operative cycle.”

Graner, an electrical engineer for tbe Sprague Safety Control & Signal Corporation, testified that abe track impulses could be completely forestalled by the engineer moving tbe brake lever “to service position and immediately back to lap position” before the automatic braking impulse starts — presumably between tbe time of tbe cab warning and of tbe braking operation. Whether this is possible is not entirely clear, but, if it is, it is an action never contemplated by tbe patentee, and is, in fact, in direct opposition to and a defeat of the entire basic theory of tbe patent, which theory is to compel tbe engineer to obey tbe signals and apply tbe brakes if be wishes to retain control. However, when tbe appellee’s device, which appellee contends is patterned after Sprague, was installed, it contained a very different and separate forestalling element, which gives pertinent reason for tbe questions of why was this done unless thought necessary and why necessary if a forestalling means already existed.

Tbe statement in tbe patent (p. 7) as to location is “in tbe cab or any other preferred position. * * * Preferably, its closure is made sufficiently inconvenient to insure its full effectiveness in case of operation” — that is, tbe location is for protection of ¿be device and not as a penalty.