to the usual defense of invalidity, the defendant urges the defense that the payment of the final fee for the issuance of the patent was not within the six months prescribed by Rev. St. § 4885 [U. S. Comp. St. 1901, p. 3382]. This defense seeks to attack the regularity of the proceedings in the Patent Office. Such attack cannot be made by a proceeding of this character. Patents regular on their face are not the subject of collateral attack. Lamprey Boiler FurnaceMouth Protector Company v. Economy Feed Water Heater Company (C. C.) 62 Fed. 590; Hallock v. Babcock Manufacturing Company (C. C.) 124 Fed. 226; Mowry v. Whitney, 14 Wall. 434, 20 L. Ed. 858. The patents in controversy relate to apparatus for recording elapsed time. These patents are used conjointly. The calculagraph of the complainant is employed for recording the interval of time between the beginning and end of an occurrence; so that the amount of time consumed in such interval can be read directly from the record. The specific use of the calculagraph is in measuring the length of time in which a long distance telephone line is employed. The evidence before us tends to show that prior to the introduction of machines by the complainant there was no complete way of timing telephone messages to determine the time. occupied and the amount of toll to be charged for the use of the telephone. The leading methods were for the operator to determine the time of commencement by observing a clock, and writing down the time with a pencil, and at the close of the period during which the telephone was in use to again observe the time. by the clock. This was a crude method, and subject to many errors. Stamps were afterwards used, which recorded the initial time and the final time, and obtained the result by subtraction. Stop watches were also used. All these methods were crude and unsatisfactory. They resulted in wasting the time of the telephone company and the money of the customer. The calculagraph, so far as it embodies the invention of the Hamilton patent, is capable of recording the amount of time during which a telephone line is in use, the record being made by inserting a card into the machine and making an imprint on it by the movement of a handle at the commencement of the period of use, and again making an imprint on the same card by the movement of another handle at the close of such period, the card in the meantime being removed from the machine, so that the initial or final imprints of other records can be made for other customers. The record, so made, can be read by simply counting graduations on the annular scale which the machine marks off, or, when numerals are used, by seeing what division number is pointed to by the pointer imprint made by the machine. By use of the Abbott patent, the calculagraph makes a record of the time of day of the initial imprint of the elapsed time. of record, the same handle being used in making the time of day record and in making the initial elapsed time record. Thus the calculagraph, by the conjoint use of the Abbott and the Hamilton patents, without adding to the number of handles, and without requiring that the card be placed in a separate machine for the purpose, is enabled to record the time of day as well as the amount of

elapsed time. It thus enables the telephone company to have a record of the time when the use of the telephone took place, as well as of the actual time employed by each customer. There is evidence in the record that the calculagraph has come into general use; that it is now employed in measuring 80 per cent. of all the long distance messages of the United States; that it is used largely in other countries, and that it has over 3,000 instruments now in operation. The general object of the Hamilton invention is stated by the patentee in his specification to be "to enable measurements of time, space, or quantity to be easily and quickly recorded by automatically changing mechanism; and to this end such. invention consists in the construction and combination of parts constituting my device."

The claim in controversy we have already quoted. It will be seen that it presents the following elements: (1) A rotatable printing die provided with (2) an annular series of progressive numerals; (3) a pointer type in juxtaposition with the zero number of the series; (4) mechanism for rotating the die at a uniform velocity; (5) type printing mechanism. The form of machine brought before the court consists of a base having a revoluble printing die mounted therein on a shaft. This shaft is driven by a clock movement, which movement is also used to drive the hands of a clock dial mounted on the base. An inked ribbon extends across the face of the printing die, and is wound at each end on spools. A bracket arm on the base supports a cylindrical barrel above the die, with its axis concentric with that on which the printing die revolves. Within the barrel. is a hollow sleeve, which carries an annular printing platen at its lower end. A second sleeve is mounted in the first sleeve, and is likewise provided with an annular printing platen on its lower end. Both sleeves are normally maintained in an elevated position by springs contained within the inner sleeve, and both are provided with handles by which they can be depressed to cause their respective platens to press a card against the ribbon over the corresponding portion of the printing die, and thus to make an imprint upon the card. In order to insure that the card be in the same position at the second imprint which it occupied during the first imprint, guides are provided for the edges of the card to engage, such guides consisting of the base of the bracket arm for one edge of the card and of a guide strip secured to the machine base for a side of the card at right angles to the card-engaging surface of the bracket base. The form of the die which the patentee, Hamilton, chose to illustrate his invention has a "pointer and a series of graduations in the form of numerals type progressively arranged, both the pointer and graduations appearing on one integral revolving printing die." In his specification the patentee says:

"It will be obvious that any desired number or order of figures may be cut upon the printing die, so as to cause its impressions to represent hours and fractions of hours, etc., instead of dollars or fractions of dollars, the principle of operation being the same in each instance. If desired, the first impression on the ticket may be of the arrow head alone and the second impression may contain the figures alone; but in such event it would be necessary that the printing die should move in the reverse direction, in order that the figures

thereon should in progressive order pass the point which was occupied by said arrow head at the time the first impression was taken."

In the embodiment of his invention which the patentee uses the pointer type is in juxtaposition with the zero numeral of the series. The type are reversed on the printing die of the Hamilton machine, as are all printing type, so that the impression may not be reversed. The outer annular platen is of such diameter as to print from the annular series of characters without printing from the pointer type. The inner platen is of such diameter also as to print from the pointer type, but not from the annular series of characters. In the operation of the form of the Hamilton machine which we have described, a card is placed against the guides at the beginning of the period of elapsed time to be measured, and the outer platen is depressed; or, if desired, both platens may be depressed; and an initial imprint is made on the card, which imprint is from the zero type and from the "series of graduations in the form of numerals. At the close of the period of elapsed time the card is again replaced against the guides, and the inner platen only is depressed, thus making a final imprint from the pointer type alone. The initial imprint, then, by showing the zero numeral, shows the position of the pointer at the beginning of the period of elapsed time because the pointer type is in juxtaposition to such zero numeral on the die. The final imprint, containing, as it does, the imprint of the pointer itself, shows the position of the pointer at the close of the period of elapsed time; and, because of the difference in radial position or appearance, or both, of the zero and of the pointer, it can at once be told which imprint was made first and which was made last; so that it can be known whether the longer or the shorter arc between the two imprints represents the period of elapsed time. The imprint of the annular series of graduations enables the length of the arc to be determined by the numerals, if numerals are used, or by simply counting the number of graduations between the initial position of the pointer or zero, and the final position thereof.

Directing our attention now to the Abbott invention, we find that the first claim embraces the following elements: (1) Dies for printing the record of intervals; (2) dies for printing the time of day; (3) a single motor for driving both sets of dies; (4) a lever adapted by alternate movements to cause impressions to be made from both sets of dies. The second claim embodies the following elements: (1) The dial and pointer dies; (2) time of day dies; (3) single lever for actuating the time of day dies and certain of the interval dies; (4) means for actuating the die or dies not actuated by said lever; (5) a single motor for driving all of said dies. The substantial difference between claims 1 and 2 of this patent lies in the addition of the element which is numbered 4 in the above enumeration, namely, means for actuating the die or dies not actuated by said lever. The invention of the Abbott patent is illustrated in the complainant's calculagraph before us. So far as claims 1 and 2 of the Abbott patent are concerned, such machine consists of dies embodying the Hamilton invention, and adapted to record elapsed time in units of

two denominations, dies for recording the time of day of the beginning of such period of elapsed time, a lever for printing from said time of day dies and from part of said elapsed time dies by alternate movements, and a lever for printing from the remainder of said elapsed time dies. In the actual exhibit of the calculagraph before us elapsed time is recorded by one set of elapsed time dies in groups of five minutes, a full revolution recording sixty minutes, while in the other set of elapsed time dies the elapsed time is recorded in units of fifteen seconds, arranged in groups of four, or one full minute, a full revolution being five minutes. In the calculagraph before us, instead of moving the platen toward the dies, the dies are moved toward the platen, the latter being stationary; and, instead of having a two-part platen, the die is made in two parts, so that the annular series of progressive numerals can be printed without the pointer at the initial imprint. The pointer can be printed alone at the final imprint. The pointer dies, while capable of separate movement from the numerals dies, always point to the zero numerals. Two levers rise from the casing of the machine; the right-hand lever, by a backward movement, operating the time of day dies, and by a forward movement operating the annular series of progressive numerals type of both sets of elapsed-time dies, while the left hand lever operates the pointer-dies alone. All the dies are driven by a single motor, and are thus kept in absolute unison. It will be seen that the Abbott invention provides an instrument for recording elapsed time. It adds to the information furnished by the machine of the Hamilton patent the time of day when the elapsed time began to run.

Certain devices in the prior art are cited by the defendant as either invalidating or essentially limiting the inventions of both the Hamilton and the Abbott patents. Foremost among these devices are the "old time stamp" and the time stamps of certain patents referred to by the defendant. Without going into the details of all the matters suggested in the prior art, it is enough to say that, while many of these devices are of great interest as showing the state of public knowledge, we do not find that any of them should properly make any limitation upon the inventions now in controversy. The thought of the patentee, Hamilton, was a device for automatically recording elapsed time. The "old time stamp" stamps upon the paper a record of the time. This stamp could be used for stamping the time when a given operation began and again. when it closed. For convenience, the stamp of the final time could be superimposed upon the stamp card with the stamp of the initial time. But there is nothing in this stamp which indicates that its purpose was to automatically record elapsed time. It certainly never could make such record. Nor do the patented stamps show any fitness for automatically recording elapsed time. In the Emerson patent the patentee describes his invention as relating "to certain improvements in that class of stamps which are employed for marking, dating, canceling, or otherwise impressing various documents, and at the same time indicating the hour and minute at which such impression is made." The Hinchman patent sets

out its object to be "to furnish a machine for hand printing or stamping, which shall, when operated, record by a single movement of a platen or impression pad the precise hour or minute of the day at which such impression was made, so as to adapt it for such purposes as stamping messages which are received or sent at telegraph stations, tickets sold at railway stations, time checks for employés of manufacturing establishments, and other analogous uses. But the automatic recording of elapsed time is not a use in any sense analogous to the use claimed by the patentee. So, in all the alleged anticipations, we do not find that there is anything to show that the vital and inventive thought of the patents in controversy has been in any way anticipated by previous patents. The inventions brought before us by the complainant represent a distinctly "original conception, so essentially unlike anything in the prior art that nothing earlier has been presented to us out of which the inventors could read the invention of the patent or their claim of it. There was no known device which could be converted into the invention of any improvement short of rejection of the entire plan." In thus commenting upon the devices cited in the prior art, we have adopted the language used by the Circuit Court of Appeals in the Fourth Circuit, in Crown Cork & Seal Co. v. Aluminum Stopper Co., 108 Fed. 861, 48 C. C. A. 72.

We now come to consider the defendant's machine. This device is called the "timometer." It comprises a base having an overhanging casing containing clockwork for driving the rotatable dies by which the time of day and elapsed time can be recorded. The machine is for the purpose of recording elapsed time in minutes as well as in seconds. The die for recording minutes of elapsed time is a circular plate having upon its surface a half circle of graduations, every fifth one of which is made longer than the others to facilitate the reading of the arc representing the period of elapsed time, and also having a pointer pointing at the place where the zero of the series of graduations would be, the zero being actually omitted. This die revolves concentric with a stationary annular time. of day die divided into hours and minutes, there being a triangular hour pointer revolving in an annular space formed between the elapsed time die and the hour die. The die for recording elapsed time in seconds consists of a circular disc having two dies formed thereon in line with each other, but of different heights. The disc is concentric with an annular dial which is graduated, every fifth graduation being elongated to facilitate reading the arc, which is the measure of the period of elapsed time. Co-operating with the dies just described are two platens-a larger one, covering all of the dies except the inner one of the two dies on the seconds disc, and a smaller platen, covering only the inner of the two dies on the seconds disc. The two platens are operated by separate cams upon a fulcrum at the base of the machine, each cam having its own lever by which it can be turned, there being provision for coupling the two levers together. The larger platen is provided with a plunger that is adapted to rest upon, and be operated by, cams formed on the inner end of the hand lever. These cams bear against