just mentioned was suggested and is carried out for his own use by a New York attorney. The blank is printed in the ordinary manner, but not cut to size, the proper dimensions being indicated by depressed lines made by an embossing die. Thus there is, while the drawing is being made, an extra width of paper all around, which is cut off easily and accurately afterwards along the embossed lines. This is a very good plan, and is recommended to dealers for imitation.

A further requirement of the office is that, between the top marginal line and the uppermost point in the drawing, there shall be a clear space of 1 inches to be used by the office for the insertion of the name of the inventor and the title of the invention. As already explained, the lower right- and left-hand corners are occupied by signatures. There remains, therefore, for the drawing itself, a comparatively small space, which one has to economize as much as possible. For this reason, the blank should be so printed that the signatures occupy as little room as possible. Fig. 1 shows a blank in which much valuable space has been wasted, while Fig. 2 shows a very good arrangement, in which the space between the signatures is available for drawing purposes, which space is lost in the arrangement shown in Fig. 1. Very often this lack of space between the signatures proves to be an annoying matter. Another difference between the two blanks will be observed in the marginal lines. Fig. 1 shows the upper and the left-hand lines thin, and the lower and right-hand ones thick; this is evidently intended to show artistic taste on the part of the printer, but the heavy uniform marginal line of Fig. 2 is to be preferred. In order to make the holes of the thumbtacks show as little as possible, the draftsman will run the tacks through the corners of the marginal lines, and if these are heavy, the holes will hardly show at all. The paper being stiff, two tacks driven as at t and t', Fig. 1, will hold the sheet firmly. To avoid tack holes, the sheet may be fastened down, as in Fig. 2; but then three tacks are needed, and the lower one will always be in the way of the T square.

The patent office designates the edge ab as the top of the drawing, and all views must, if possible, conform to this idea. In cases where the length of the figure is greater than its height, the sheet is turned around so that ea is the top of the drawing, the signatures being then to the left. To this rule of the office it is well to add another; namely,

whenever possible, all the sheets of one set of drawings should read the same way. This has the advantage that, when laying out the several views, and also when studying the drawings, the sheets can be handled with much more convenience by draftsman, attorney, and official examiner.

Although a patent-office drawing is mainly a picture, or illustration, it will pay the draftsman to work to scale, using the same scale for all views belonging together. This will not only greatly facilitate and expedite matters, but will make the drawings a good deal clearer, and the examiner will be able to see at once the relation of the various views to one another.

Suppose, for instance, that a certain machine is to be illustrated. We find that to properly show the complete mechanism

FIG. 3.

it will require a plan, a front view, a side view, and a longitudinal section of the whole machine, as well as various details. The draftsman may then tack three sheets on a large board, as in Fig. 3, and proceed as if he were making a regular working drawing. He will thus commence and finish all the three main views at one time. It is evident that the scale should be as large as possible. The details will naturally be drawn to a larger scale than the main views. By using the same scale for all the details, the reader of the drawings will much more easily see how the different parts of the machine fit together.

There is a great difference between the details of a patent drawing and those of a working drawing. In the latter the details consist, as a rule, of one or more views of each single piece of the machine, while in the former the details are intended to convey to the reader a more lucid conception of the operation of one particular part of the mechanism. Thus, if it were desired to make clear the forming of the loops in a sewing machine, two or more detail views of the shuttle and needle bar might be necessary, showing these parts in various successive relative positions towards each other. In

arranging the views one should strive to keep the various parts of the machine belonging together as much as possible on the same sheet. Thus, in the given example, the loopforming mechanism should be shown-in as many views as necessary-on one sheet, and the cloth-feeding mechanism on anotherthat is, if both cannot be put on one sheet. When it was said a little while ago that the comparatively small space allotted should be economized, it was not meant that the views should be crowded. On the contrary, they should be kept well apart. In fact, the office requires that each view be distinctly separable at a glance from the others. Thus, an arrangement of views as indicated in Fig. 4, in which the plan cuts, as it were, the elevation in half, is not permissible. There is good reason for this: Every letters patent that is granted is made public through the official "Patent Office Gazette" by a single insertion of the claims and one illustration. This illustration is chosen from among the complete set of drawings, as, in the opinion of the examiner, being best suited to illustrate the invention. If, now, the views are so interspersed that one cannot be photolithographed without the other except by great trouble, there is more space needed in the "Gazette," than one patent is entitled to. If it is possible to leave a little blank space on each sheet without making it look too bare, it is well to do so for the possible addition of a small detail view here and there.

In regard to the selection of the various views, it is evident that no fixed rules can be given, as it will depend on the subject to be illustrated. Generally, however, the draftsman will have among the figures one that represents the whole invention, in a general way, as completely as is possible in a single view; in the case of a machine, this will probably be an elevation, taken from that side which shows the most of the mechanism. This main view generally constitutes Fig. 1, and serves a double purpose; firstly, it is intended to be the one to be published in the "Gazette," showing the subject of the invention in a general way without going into details; secondly, it is a very great help to the attorney, as well as to the examiner in the patent office. With the former it is the basis upon which to build his specification, and to which to refer when discussing the details. With the latter it forms the clue that at once places him in touch with the gist of the invention, and saves him the trouble of reading the whole specification to learn what it is all about.

The fact is, the set of drawings should convey to the examiner a full understanding of the operations of a machine, without the specification. This is evidently what the office expects, for candidates for the position of examiner are given a set of drawings, and requested, as a proof of their qualification for the place, to briefly state the nature of the invention from the drawings alone. That a candidate who is unfortunate enough to be given a poor set of drawings will make also a very poor guess, is evident. In the "Rules of Practice," we find the following sentence: "When the invention

consists of an improvement on an old machine, the drawing must exhibit, in one or more views, the invention itself, disconnected from the old structure, and also, in another view, so much only of the old structure as will suffice to show the connection of the invention therewith." As a rule, this requirement cannot be strictly adhered to. Such improvements will generally be so intermingled with parts of the old machine scattered through it that a separation, as called for in the above rule, would make a drawing incomprehensible. A good way to separate new parts from old ones is to "shade up" the former only, and show the latter in outline, or even in dotted lines, as in Fig. 5.

views, if not very skilfully done, have always a more or less awkward and unnatural appearance, as in Fig. 6, for instance. This is especially true when approximate perspective methods are resorted to, such as "cavalier perspective" and "isometric" drawing. As a rule, then, the draftsman will employ any of these methods as sparingly as he can, and only where they offer a decided advantage, as, for instance, when a single perspective view will more clearly demonstrate certain points than a number of rectilinear projections would (as in Fig. 7, for instance), or in cases where the shape of an object plays an important part, such as in collar buttons, garter hooks, shoes, etc., or as in Fig. 8, which represents a belt fastener. In this matter the draftsman must follow his own judgment.

In shading and in applying shade lines, the light is supposed to come from the upper left-hand corner at an angle of 45°, as is customary in most mechanical drawings. This applies to all the views. The patent office requires all lines, no matter how thin, to be perfectly black and sharp. This calls for the best kind of ink and the very best instruments, kept in perfect order.

Draftsmen engaged in illustrating work prefer generally to grind their own ink from a stick of Chinese, or India, ink. A thick slab of slate into which a dish has been turned, and having a still lower and smaller cavity at the bottom, makes a good ink well, and a glass cover keeps out the dust. In such a well (see Fig. 9) the ink keeps a long time

and remains clean. Slate is much to be preferred to china for ink slabs, as the stick can be ground much faster on it. Tepid water accelerates the process of ink rubbing. The commercial liquid inks are less suitable for patent-office work, although they are much

used. The best of these liquid inks is Higgins's American Drawing Ink, of which there are two brands, namely, "general" and "waterproof." The latter is not so intensely black as the other, does not flow as easily, and is more troublesome for the pen. The "general" ink is, therefore, to be preferred for fine work.

With respect to instruments, every draftsman will have his own way. In selecting a ruling pen, however, one will do well to choose a "duplex," as shown in Fig. 10. This is adjusted for the thin, unshaded outlines of the drawing by the screw a, and for the shaded outlines by a screw b. By pressing lever c against the end of screw b, the pen opens for a shade line; by letting go the lever, it closes on screw a

valuable, if for no other reason than this, that all dust and ink or pencil marks are easily detected and cleaned away before soiling the white sheet. Otherwise, hard-rubber blades and triangles would answer as well-the flexibility being common to both materials. It will be evident from the above that squares and triangles of wood with transparent celluloid edges, which were offered by the trade some time ago, are of no earthly use. For penciling on bristol-board a very hard pencil should be used-Faber's HHHHHH,

We take it for granted that the reader knows that patent-office drawings are reproduced by the lithographic process, and that, therefore, they must be executed with the pen only; that is to say, no brush washes are permitted. Draftsmen who have not had a great deal of experience with freehand work and lettering will find some difficulty in handling the common pen for shading and lettering. This is purely a matter of practice.

FIG. 9.

Some artists are able to do with ease as fine work with a common pointed writing pen as others with the finest ruling pen. But these are exceptions. For lettering and freehand

FIG. 10.

GAS

be clearer. Thus, if in a section there are two or more superimposed cylindrical surfaces, he may leave out the shading on one or two of them [see Fig. 12, (a) and (b)]. View (b) does not look artistic, but is a great deal clearer than (a). As a matter of fact, in cases like Fig. 12, the shading of the concave

uneven section lining. A drawing thus faulty is at once a telltale on the beginner and amateur.

Under the same head come a good many other things which will distinguish the work of an experienced man from that of a beginner or a poor workman. Here are a few instances. No one who has ever seen a gear-wheel draws the teeth as in Fig. 14 (a), but makes them look shipshape, as in (b). They should be made with the compass and correctly spaced. Rather than indicate them either faulty in shape as at (a), or carelessly by hand as at

cylindrical surface c is almost invariably left off. In the same manner plane surfaces are usually not shaded, except in perspective views, and in cases where several of them are superimposed, and where clearness demands that the surfaces be distinctly separated [see Fig. 13, (a) and (b)]. In such a case, equidistant thin lines are used, either horizontal or vertical. Section lining is, as usual, made with oblique parallel and equidistant lines as open as the case will permit.

The plane upon which a sectional view is taken should be indicated on the general

(a)

(b)