The Repairing & Restoration of Violins Part 3

Another method has been known as the French, and when neatly done is one of the most sparing of the old wood (see diagram 24), but it is beaten in this respect by another foreign method (diagram 26), which is less evident to the eye, although requiring more skill in accurate cutting and adjustment. Another yet more secret I have only seen in an Italian grafting, and it may be native; no join whatever is seen in a front view nor in the peg-box if this part is at all soiled or dusty, as is usually the case. This is owing to the join--there must be one of course--being each side at the angles formed by the walls of the peg-box. This is counterbalanced however by the necessary cutting away of the central line or ridge at the back for a considerable distance. If done accurately and artistically, all very well, but this is not likely to be always the case, although a comparatively easy bit of work with the original lines each side as a guide. This method of grafting is puzzling when successful, as little or nothing is perceptible from the front and not much, unless searched for, at the back.

On measuring the different parts of an old violin in its original condition, we shall find the neck, taking from the edge of the upper shoulder of the instrument to a point where the nut is placed, to be not much more than four and a half inches, whereas our modern necks measured at the same parts would give five inches and an eighth. The old length taken at this part alone would give too short a fingerboard, causing the fingers to hamper each other, especially in the upper part of the register, where so many modern composers seek for effective pa.s.sages. The neck must, therefore, to meet the requirements of the day, be lengthened. In the earlier part of the present century there was a method much in vogue for effecting this without interfering with the head and while keeping the greater part of the original neck (diagram 23). This was done by firstly removing the fingerboard, probably worn into ruts; the middle of the thickest or lowest part of the neck attached to the body had then to be loosened and removed, often no doubt a troublesome task owing to one, two and sometimes three nails being in the way, this in consequence of the habit of the old makers of attaching the neck with its scroll before closing up the body of the violin. Having accomplished this, the repairer chiselled off two square pieces, one on each side at the same end, and then fitted longer blocks with the grain running the same way. These were afterwards cut down to the proper form, so that the terminating part under the fingerboard increased the length of the neck to the modern standard.

Of course, when fitted into the original s.p.a.ce or socket from whence the neck was taken, the rounded part going to or above the b.u.t.ton was now too large, this part was therefore cut, filed and finished down to the required size and shape.

This method of lengthening the neck, however, went out of fashion as connoisseurs and performers, finding the old necks so frequently devoid of figure--the reason being probably that plain wood answered best for the cutting of the volutes--made the repairers remove the whole of the neck and subst.i.tute one of the best figure they could obtain.

This forced fresh attention to the splicing as it is termed of the scroll to the neck or graft, and the method has continued to the present time of clearing away the whole of the neck and using handsome wood.

Further impulse was given to the practice by the fact of the fingerboards put by the old makers rising so little above the body of the instrument. The bridge was made very low to accommodate this state of things. The increased rapidity of the movements of the bow from one string to another over the middle ones in the performance of modern music made a higher one absolutely imperative, as the heel of the bow would too frequently chip pieces from the waist curves. There were thus three good reasons at least for placing an entirely new neck on an old violin; firstly the plain wood of the original maker, shortness and the low angle with regard to the plane of the body.

In order, therefore, that everything may be accomplished with sufficient exact.i.tude, we will begin with the roughly sawn block ready for measuring and shaping up for its destined purpose. The scroll, which is to be replaced on a neck according with modern ideas, we will suppose to be on an Italian violin that has come down to us from the early part of the last century. The violin tuned up to the present concert pitch and music of our period having many of the modern style of difficulties, would prove utterly inadequate to the task of giving out its tones in a manner expected of it.

In proceeding to work then, the workman executing this modification having selected his block of curled maple, planes it to an oblong of equal breadth. He calculates as to the best position for showing off the curl on each side of the neck when finished. Having decided which is to be the upper part or that covered by the fingerboard, this is planed to a good level and smoothed. A line drawn with a good pointed pencil or pointed knife, and sharply defined, is then drawn down the whole length exactly in the centre. At the end which is intended for the thickest to be inserted in the body of the instrument an equal width each side must be marked. Near the other end, at a distance that shall correspond with the opening of the peg-box, an equal width each side of the line must be marked off as at the other end. These two measurements will represent as nearly as possible the width of the neck along its course at the junction of the fingerboard. From the point representing the opening of the peg-box one of two lengths upwards must be decided upon; if the splicing is to be effected in the manner common in this country, a greater length will be required than for that of the French style. This latter is more to my fancy than the other, as there is less of the original wood lost. If for the former, a length of wood beyond the opening will be required of two inches, if for the latter or French a little over one inch and a half will be enough. The central line has of course been continued for the whole length of the wood. The waste wood at the end can now be sawn off down to the line.

The next measurement will be, supposing the French style is adopted--that of the extreme width of the end, which will be given by taking a point at half the thickness of the peg-box wall at the part and similarly placed on the other or opposite and taking the width between the two. This divided equally and marked on the wood of the new graft each side of the central line will give the narrowest width of the part to be inserted in the peg-box. The outside may be then removed by the saw vertically. There will now be necessary the marking off a part on the graft that shall represent the thickness of the nut or the distance between the end of the fingerboard and the peg-box opening; the breadth across, or we may call it the length of the upper part of the nut, will be exactly that of and at the part where the opening will be made in the peg-box for the reception of the graft.

The wood to be cleared away outside the lines which mark the width along the course of the fingerboard will be the next proceeding; it may be done neatly with a rather fine toothed saw and then carefully planed up closer to the lines, barely touching them. It is preferable to leave the sides for the present at right angles with the top surface, although they will not be kept so for long, but by thus working the measurements are facilitated. Going to the lower or wider end a line must be accurately marked quite square with the long central line, if not accurate the whole work will be thrown out of truth. On the sides there may now be marked and roughly sawn away (diagram 28) so much of the wood that shall leave enough for the cylindrical part that is to be finally rounded and finished off for handling. Care must be taken that the rounding commences underneath, a little away from the part that will be fitted into the peg-box. This of course must be according to measurement or template kept for the purpose if graftings are likely to be wanted in the future. There will now be required the marking of the exact form of the part that is to be inserted in the body of the violin, or more strictly speaking, into the upper block. This is done easily from a pattern cut to shape and size for instruments of average proportions. Sometimes, owing to the height of the ribs, the pattern cannot be applied so as to fit; in that case fresh lines must be drawn to measurement as with the central one on the fingerboard plane.

[Ill.u.s.tration: DIAGRAM 28.]

The line dividing the part exactly in the middle must be accurately done, the distance at the narrowest or lowest part that is to be glued on to the b.u.t.ton carefully marked, allowing the top part when placed in position to be a quarter of an inch above the border (diagram 29).

The width of the lowest portion must be mainly guided by the size of the b.u.t.ton, which, although there is an average of a rough kind, is sometimes small, at others very wide. The width must be taken of the b.u.t.ton, carefully divided into two equal parts to be marked on each side of the central vertical line (diagram 30). All below what is necessary to keep may now be cut away, the surface being kept parallel with the fingerboard plane. The parts outside the slanting lines may be hewn away, the surface running evenly with the outer lines of the fingerboard width so far as it extends, which will not be more than about an inch.

[Ill.u.s.tration: DIAGRAM 29.]

[Ill.u.s.tration: DIAGRAM 30.]

The next process will be that of excavating the part that is to receive the root or end of the neck. If the instrument has been accurately constructed with the join running down precisely in the centre, the line already marked on the root of the neck will be a safe guide for marking each side of the join the width of the portion to be cut away.

The depth inward of the cutting should be an average of a quarter of an inch. In case the already excavated part in an old and much repaired instrument is roughly torn about and made unequal in its measurements, attention must be fairly directed to this part separately; that is, if too much wood has been cut away on one side it must be replaced by fresh, after clearing away irregularities in order that a good fit may be accomplished. The fresh wood must be neatly inserted or placed in position and may be held in position during the hardening of the glue by supports or wedges placed across from side to side. When quite fit by reason of its dryness, the distance from the centre must be marked and the fresh wood cut away to the required depth and width with a keen edged chisel and small shavings cut at a stroke, as there will be some cutting against the grain to be done besides working in a confined position.

Great regard must be paid during the process of cutting this part that the corners or angles are quite cleared out, or the neck when inserted as a trial or rehearsal will not give a truthful report of the accuracy of the incisions owing to some insignificant portions sticking up and causing the neck to look awry. So far we may take the fitting as having been accurately done to the central line down the middle of the instrument; but now comes a further process in connection with the adjustment of the neck, and that is, the rise and inclination of the level of the fingerboard in relation to the bridge which is to be fitted eventually.

The average--it may almost be called the standard--height of the violin bridge is one three-eighth inches. There are occasions when this measurement may be departed from, as in the instance of a high model, when an eighth, or even more, may be taken off with advantage. This must not be taken as necessary for the proper emission of the tone from a highly built instrument. The raising of the bridge in modern times is due to other causes, the most important being that of allowing better play or room for bowing rapidly over from side to side without rasping the border at the waist. It is an alteration which accompanied the lengthening of the neck and stop in the early part of the last century.

To obtain a proper setting or inclination of the neck, several ways are adopted by repairers. They vary according to the kind of guide or pattern used. This is usually cut from a piece of hard wood, sycamore or pear. It is sometimes made as a double guide in the adjustment of both the inclination and the elevation of the under part of the fingerboard above the body of the instrument at the junction of the neck. This we will call No. 1. The other, No. 2, is similar, but has the height of the bridge only as a fixture, the rise of the end of the neck above the border being higher and lower at discretion. Another way, No. 3, is somewhat like No. 1, but would be used when the fingerboard is glued on before the setting. This should only be done by a fitter of some experience, as a little error in calculation is likely to lead to disaster. With each the application of the guide, or pattern, is the same, namely the testing of the exact coincidence of the inclination of the top surface of the bed under the fingerboard (diagram 31), or above the latter as shown in diagram 32. While getting the right inclination, in both instances it will be found necessary to ease the fitting of the neck into its socket, as the difference of the angle at which the neck is inserted causes an increase in the tightness of the contact of the parts. The lower part of the facing that is to be inserted in the socket, will have to be made to go into it at an angle conformable to that of the inclination or set of the neck. This will require executing with precision, and great care will have to be exercised that the squareness or rectangular disposition of the upper part already fitted and adjusted to the middle line down the instrument is not interfered with. It will be well to test this as the work proceeds. Some of the lower part, that coming into contact with the b.u.t.ton, will have to come away in order that all parts may fit, and when fixed, form a h.o.m.ogeneous rigid part of the structure.

[Ill.u.s.tration: DIAGRAM 31.]

[Ill.u.s.tration: DIAGRAM 32.]

To ensure complete success in all the foregoing operations, every tool in use must be well sharpened, and all the guide lines accurately drawn.

Part of the neck left rough and projecting beyond the b.u.t.ton may be left for future manipulation, but the joints that are to receive glue--if done by a workman of skill and experience--will fit almost air tight. With regard to this, the parts likely to give the amateur most trouble will be the exact fitting of the flat opposing surfaces of the root or squared end of the neck or graft with the socket. It will be necessary to get a perfectly flat surface. In the first place, gla.s.s papering must be avoided, not from unsuitableness of material, for in that respect it is a temptation, but from the difficulty of regulating the pressure of the hand; with the exercise of the utmost care in handling the gla.s.s paper, even when it is backed by a piece of hard wood, there will be found, when the test comes, a rotundity of surface that was deemed impossible under the circ.u.mstances. Careful sc.r.a.ping of the surfaces must be chiefly relied on for exactness. As good a mode of proceeding as any is as follows--after the first roughing into shape and then flat chiselling has been done to as great a nicety as possible, all the irregularities--there are sure to be some--can be worked down with the edge of a straight square file, used very steadily and crossways repeatedly. This done sufficiently and tested with some hard and truly cut substance, metal preferably, will be an exactly flat surface for working upon the final or finished surface.

The next thing used will be a carefully sharpened and keen edged steel sc.r.a.per. To put this tool into proper order for the purpose, it must be sharpened on a hone, not exactly at right angles, as the first impulse would suggest. The hone, or stone, must be quite flat and unworn.

If done carefully, a nice level edge will be perceptible along its course, but it is not yet at its best. Placing it on a bed of hard wood, or evenly shaped ma.s.s of iron, projecting over and held firmly in position, a good stout brad-awl may be pa.s.sed along from end to end, keeping the awl perfectly flat on the horizontal surface. The sc.r.a.per may now be turned over and process repeated, but not in the same manner or angle, for the awl will be held vertically with the handle downwards and firmly pressed along the edge at right angles with the horizontal plane, this will cause a burr right along which will have a razor-like sharpness and cutting power. This sc.r.a.per can now be applied (not too heavily) over the filed down surface, and thus work down finally all irregularities left by the file. The adaptation of this tool will at once be perceptible in the fine whitish soft shavings that will come off during its application. A little repet.i.tion across and across should give an almost perfect level. Different sized sc.r.a.pers may be used for the other surfaces where it is desirable to obtain the most accurate fitting. If all the processes have been properly carried out, the parts when tried by inserting the neck or graft for trial, should fit together without the slightest looseness or wriggling. They will now be ready for permanently glueing together.

It will be seen after perusing the above that the fitting of the neck or graft to an old violin, well or badly preserved, is a task not to be overcome easily and satisfactorily without much care and no little practical experience.

The whole of the work must be well looked over and tested as to the accurate fitting in all respects; nothing must be left uncertain or loose; in fact the flat surfaces for a perfectly successful result at the junction of all the parts should be as nearly as possible air-tight.

Having seen that the glue of good quality is strong and clean, the surfaces that are to come in contact may be brushed over with it. For this purpose a small hog-hair brush of about three-eighths of an inch wide is handy. Where the grain or threads of the wood run parallel with the surface--this being less absorbent than the other parts,--there will be less painting over required, but where the grain comes end upwards to the surface the glue will be rapidly absorbed. The painting over these parts must be repeated, the glue as a matter of course being kept warm--all the work ditto, until absorption ceases. This is a matter of some importance, as in many instances joints have become loose or broken apart, not from the perishing of the glue or damp, but from the want of this precaution on the part of the repairer during this preliminary proceeding. It must be borne in mind that this is to be a permanent junction, not to come apart from any jarring or rough usage; it is also to be one of the most rigid, and only to be separated by a saw or chisel in the hands of some future repairer when it shall be absolutely necessary. Sufficient glue having been applied the work may be put aside.

It will now be apparent to those who have possibly done a little glueing that the whole of the wetted portions have to some degree swollen, and therefore if the junctions were brought together they would be found too tight and refuse to meet. Just so; and that is one of the reasons for placing the work aside until the glue has dried at all the parts painted with it. When after a sufficient time has elapsed the work is examined, it will be found to have contracted to its old size and form.

There will not be the necessity for waiting till the glue is at its hardest; probably time will not allow of this, some days being absolutely required, but it must, for the next process, be very firm and seemingly dried through.

Examination will reveal the fact of the whole of the surfaces that have been wetted, being raised or roughened under the foregoing operations.

Recourse must now be had to our small sc.r.a.pers again. These will again be applied carefully across and across the surfaces, until, in the judgment of the operator, the surfaces are level and clean. Particular care must be taken with the edges, angles and corners, that the superabundant glue is removed. The right angle of the sc.r.a.per will be used for this purpose, and should a small particle or two, at any angle or corner, refuse to budge at the request of such a light tool as a sc.r.a.per, the powers of a sharp chisel must be brought to bear upon the subject, and the obstacle removed. Close attention should be paid to the above, as in the operation of first glueing, the wood, or woods, having unequal absorbing powers, will swell in accordance therewith, and upset the calculations that have been so carefully made for the close junction of the parts. For first rate work, the sc.r.a.ping must be so carefully and accurately done down to the surface of the different parts, as to leave little or no glue above the surface of the wood.

The desired result will thus be secured, that of the pores of the wood being closed or filled up. For the next stage the glue will not be necessarily quite so strong, a degree weaker will do. Everything must be ready to hand, including a cramp of sufficient size and strength.

Before proceeding further, however, the manner of the application of the cramp must be considered. And now, how are we to obtain a direct pressure of the cramp on the largest surface, and which would have to be in one direction, end to end of the violin, seemingly a perfectly impracticable matter? The answer is, it is not practicable, hence the above numerous injunctions as to preliminaries, and which have to do with counterbalancing the impossibility of direct and strong pressure.

The only pressure that can be applied directly is that of a nearly perpendicular one, the cramp grasping the b.u.t.ton from underneath, with a proper guard or padding of millboard, or cork, cut into shape.

A full brush of glue will now be pa.s.sed over the whole of the surface of the socket, or receptacle for the root of the neck; which latter must an instant after be treated in the same manner. The two must be treated as one operation, and in a warm atmosphere. In the summer time no extra precautions will be necessary; but in cool weather the strong glue will soon set if the parts to be operated on are not kept in close proximity to a heating stove, or fireplace, or the apartment kept at summer heat. The neck and socket being thus kept at a warm temperature, the former will be firmly thrust into position, and with hand pressure put as close as possible. The superfluous glue will ooze out all around at the junction of the different parts; if it does not, that will be a sign that there is a looseness somewhere, or the surfaces have not been forced together close enough. This must be seen to at once, the parts separated and examined. It may be that the failure has happened through carelessness in allowing a chip to get in, or a piece of grit has prevented the opposing surfaces coming together. This being removed by a small knife, the brush, with a little more fresh glue, may be pa.s.sed over the surfaces again, and the fitting this time should be perfect. The cramp and padding should be at once placed in position and screwed down tightly. All glue appearing above the joints should be carefully wiped away with a cloth kept ready to hand for the purpose; it is better to do that now than have to sc.r.a.pe or cut it away when hard; it will also save time.

Ample time must be allowed for the glueing to thoroughly dry. This must be estimated according to the conditions of the time and place. In very warm weather, or where the atmosphere is heated artificially, the time consumed in the drying and hardening is less than when the air is saturated with moisture.

When on examination the dryness is such as will warrant the removal of the cramp, this can be done. If all the measurements, fitting and precautions have been duly attended to, the neck or graft, with its line in the centre--supposing the present method is that adopted before the fingerboard is placed in position--will form an exact continuation of a line down the centre of the violin. A look down from end to end, or placing a long straight edged rule against the line, will be a way of testing this: if all is correct, the line will be perfectly straight and not bent. Should the latter be the case, the measurements, or fitting, in some respects, will have been inaccurate. It would be very provoking to find it so after all the trouble undertaken, and many instances are to be seen where the work has been left in this condition, and the stringing up and regulation has been, not only under great disadvantages, but absence of comfort in playing, and indeed the proper emission of the tone has been sacrificed. If the violin is one that is worthy of being performed upon with skill, there is only one alternative to putting it aside as useless, that of having the neck sawn off and the whole process of renewal gone through, with the aim of next time being more careful and true.

Supposing, however, the neck is truly set and all is satisfactory, the next stage will be the laying of the fingerboard. This should be of good, close and straight grained ebony, free from knots. Fingerboards are usually sold in the rough; that is, with the upper surface, or rounded part trimmed down to an approximate curve. They are cut to lengths of about ten inches and a half to three quarters. Should the violin require a fingerboard less in length than this, a small portion must be sawn off, preferably from the small end. Great care must be exercised that it is done in right angles with a central line drawn from end to end. As the drawing of this line would entail some trouble, the under, or flat surface can be placed face to face with one that is known to be quite true, and a line with a fine pointed pencil made, or better, a scratch with the point of a small knife, guided by the true end of the perfect one. A fine toothed and sharp saw will remove the unnecessary wood. In doing so, precautions must be taken against splintering and spoiling the wood. To prevent this, a piece of waste wood, cut slightly out of the square, should be placed against the stop of the bench, so that when the ebony is placed against it, the sawing can be done flush with the side of the bench. The saw should be fine, in good condition, and gently used, or the line made will be ragged, ebony being brittle and splintering stuff, requiring some humouring in this respect. If the sawing is accomplished neatly and vertically true--this last is very essential--there will be little to do in tr.i.m.m.i.n.g the surface of the end that is to come against the nut when near completion. A piece of fine gla.s.spaper wrapped round a squared piece of pine, will make a good surface. The reduction of the width of the fingerboard at each end will then be proceeded with. In the case of an old neck being retained, the width of it at each end can be taken by compa.s.ses and marked on the flat side of the ebony. A thin shaving should be allowed for in finishing off. But we are on the work of a new neck; therefore the marking off should be done to some general standard. A good one may be reckoned as follows, for a violin of fourteen inches long and average width--total length of fingerboard, exclusive of nut, ten and a half inches--greatest width, one inch and five-eighths, width at nut, one-sixteenth under an inch. The ebony will be planed neatly down, with vertical sides, to these measurements. The height, or rise of the sides of the fingerboard above the maple, three-sixteenths of an inch, which may be kept for the whole length.

The reducing to the requisite width and depth should be done with the plane in good order, a metal one for this kind of work being the best.

The surfaces that are to be glued together must now be considered. An untidy looking black line along the neck at the junction of the ebony and maple goes far to spoil the general effect; a glance at this part will at once be sufficient for declaring whether the neck and fingerboard has been fitted by a neat and competent repairer. A frequent cause of the dark line--it is really a want of proper fitting together of the parts--is the hastily planing the two surfaces--straight enough possibly--and delay while the glueing operation is in progress. The fact of ebony being almost equally affected by moisture as other woods--in fact, more so than some--must not be lost sight of. Coupled with this curling of the wood under the influence of damp is the want of proper regulation of the pressure after glueing and placing the parts in opposition. An old-fashioned method of uniting these parts is still pursued by some repairers--the surfaces are planed evenly, the glue is applied over them, they are clapped together and string tied tightly as possible. Diagram 7 will show a modern and improved method, that of a mould of soft wood for back and front of neck and fingerboard. In affixing the fingerboard many repairers have left a gouged channel reaching from the nut to the end or insertion of the neck. This may be seen sometimes on turning the part towards the light. The intention seems to be from an economical view, that of removing the ebony, if necessary, without injuring the glued surfaces by pouring a little water down the pa.s.sage and waiting till the damp enables the fingerboard to be pulled off without fracture.

This tedious operation is wholly unnecessary, for the time spent would be worth more than a new one with its tr.i.m.m.i.n.g up. Some repairers have used a toothed plane on the level surfaces to enable the glue to grip well. This is another mistaken idea. The fingerboard should not be treated as a permanent part of the structure never to come undone, it should be so secured as to last as long as required under fair usage, but in case of violence it is best that it should snap clear from the neck than hold tight enough to distribute, or concentrate, the strain on other and more delicate parts of the structure. Experience has suggested the following as generally best for all practicable purposes.

The surfaces having been made true under the plane--this should be tested before the parts are glued, when, if true, there will be no line or the very faintest one seen. Along the middle of the ebony a very shallow gouged channel may be made, about half-an-inch wide and just deep enough to prevent the glue from touching when the fingerboard is placed in position. The level across over this channel from side to side can be tested by a metal straight edge or truly trimmed sc.r.a.per.

Occasionally from damp or the action of the plane the surfaces of both maple and ebony become slightly arched; in reducing this the sc.r.a.per may be used with good effect, and a smaller one to take the least shaving more off near the channel, the even pressure when applied will close the outer edges more effectually.

The glue to be used under present circ.u.mstances should not be strong, and if the atmospheric temperature is below sixty, or perhaps not down to that, the surfaces of fingerboard and neck should be warmed. When all is ready, see that the ebony is placed evenly in the centre and then proceed to apply the cramps in the manner before described (diagrams 33 and 34). The one placed over the b.u.t.ton and the arch of the fingerboard in opposition to it must be sufficiently large, and the hollowed soft wood mould, or pad, should be more highly arched than the fingerboard, so that when pressed down, the outer edges, and not the centre of the latter, should receive the greatest pressure. The other cramps having been screwed down under the same conditions, the work can be placed aside to dry and harden. When a reasonable time has elapsed, according to atmospheric conditions, as in cold, damp weather, more time should be allowed, but under all circ.u.mstances the most dry and sufficiently warm locality should be chosen; the cramps may be removed, and of course the moulds too.

[Ill.u.s.tration: DIAGRAM 33.]

[Ill.u.s.tration: DIAGRAM 34.]

We must now see to the working down of the graft or neck, not only to the requisite dimensions, but for the finish with some effort at style.

By this last is meant such attention to evenness of contour from the b.u.t.ton along to the edging of the sh.e.l.l, as shall be strong enough without looking heavy or clumsy; much of the nice appearance of this portion of the work depends upon the neatness of the workman. a.s.suming the b.u.t.ton to be of the normal standard, or we may say, well calculated with regard to size for good effect--a good average width of this at the base where the curved line springs from the border is thirteen-sixteenths, and the projection forward--as it is not a geometrical curve--a half an inch. Some of the old Italian makers left the b.u.t.ton very large, others small. The latter never pleases the eye of the connoisseur, who, accustomed to the proportion given it by the best masters (also the modern makers), thinks it looks poor and incomplete. As the neck or graft has. .h.i.therto been left but roughly hewn out, it will be projecting for some little distance beyond that which would be occupied by any b.u.t.ton of average dimensions. Supposing the b.u.t.ton to be too small, or injured, and an addition necessary, for the next move we shall require a pair of compa.s.ses; with these, after finding a centre of the segment of a circle formed by the outer edge of the b.u.t.ton, with the other point find the distance inward, or the proper segment that is to be retained, bearing in mind that every possible part of the original b.u.t.ton should be preserved; putting the point of the compa.s.s on the central spot as lightly as circ.u.mstances will allow, a thin scratched line must be made with the other point.

Here we may remark that screw compa.s.ses should be used, so that distance apart of the feet may be kept rigid, as the width and the same circular scratch marked on the b.u.t.ton must now be marked on a piece of maple, or, as is sometimes done, on ebony. The grain of either should run in the same direction as that of the b.u.t.ton, or if done aslant it will look ugly when worn a little. The materials being matched, the fresh piece should be reduced in thickness to very little over that required for the height of the edging for the b.u.t.ton; it should be a small cake of wood large enough to cover and leave a margin where it is to be fitted.

The wood of the b.u.t.ton outside of the scratched line will now be cut away down to the raw wood of the new graft; it must be done with small and sharp chisels, carefully paring it down, leaving the edge up to the scratched line quite perpendicular and smooth all round: the strictest attention to this is necessary, so that the fitting on of the fresh edging may be done with exactness. We now take up the little cake of ebony, or maple, as selected, with the scratched line, which should be made from a central point close to the edge as possible: if this is a difficulty, the centre may be taken further in, the circular line also and the superfluous wood cut away to the central point, but not in a way to interfere with the equipoise when the edging is placed on for fitting; if this is not seen to, the edging will, when finished, look awry. The middle within the scratched line may now be first gouged away and the wood cut with a sharp knife close up to and at exact right angles with the plane. A rounded file may with care be used to make a more even surface or run. If all this is done with precision, the parts may be tried together for testing--the glueing may be seen to and the cramping done, with care that the fresh portion does not slip during the process out of its place. Some repairers would be tempted to rely upon the exact fitting, and simply slide the parts together, squeezing them well, but this is always risky. The work may now be put aside to dry.

The next proceeding will be that of working down or levelling both the ebony fingerboard and the graft, or neck. The first, in the state usually sold, will have an apparently well-finished off arching that may sometimes be near enough for letting alone after a little polishing down, but as a general rule it is not so, and further, if having been long "in stock," it may have settled down a little out of the straight during the seasoning process. Recourse should be had to the plane, a rather small metal one in good order with a keen edged iron. This must be closely regulated, or the surface worked upon will not be even but torn; hard woods, as before observed, require humouring and working down gently. The exact arching--in good work--of the ebony should be governed by a cut mould, one for each end. These may be made of some hard wood or metal, such as zinc, and if truly made will last any length of time. They should be trimmed to fit some fingerboard that has been ascertained to be just the thing in its arching. It may be as well to observe that some violinists prefer using a rather flatter fingerboard than others, but the medium is without doubt the best, and is not difficult to arrive at. The plane must be gently worked along from end to end of the fingerboard with as little pressure as possible,--hence the careful regulation and sharpness,--or you will find after a short time that instead of a nice even line, which must be tested from time to time by a straight edged rule, there will be a curved one, and this will necessitate further working down to the danger of losing thickness and sufficient strength in the ebony. If attention is paid to this, and a satisfactory even run of surface is obtained, gla.s.s-paper on a piece of straight, soft wood, but not of the finest degree, will be suitable for the present.

We now return to the modelling of the neck from the lower part of the back of the scroll down to the b.u.t.ton. This last, with its fresh edging or shield, will require another scratched line, making two semi-circular ones; it must be done from the same centre and calculated to allow of the wood being hewed away outside, leaving the full measurement when finished off.

A chisel will now be brought into requisition for removing the useless wood outside the line last marked. The cushion or sandbag must be brought into use, the violin put face downwards, the fingerboard resting in a hollow. The neck or most convenient part for holding the whole with firmness must be held tightly, the chisel then worked downwards from the b.u.t.ton, but not too far so as to cut into the portion that is to gradually enlarge, or form the quarter of a circle or the thickest part of the neck.

We shall now use a strong coa.r.s.e wood file and turn the instrument round and about, work away the neck until just outside of what will be left when the polishing down takes place.

The curves should all be balanced well even while in the rough and the contour viewed from all points should be regular. The other end of the graft will require the same kind of attention, care being taken that too much wood is not removed. The level from each end must be seen to, leaving just sufficient wood to allow for fining down; the proportions must be well calculated, thus the upper end under the nut will be hewn down thinner than the part approaching the b.u.t.ton, the line from each end being made quite even and the curving of the semi-circular shaft gradually tapering upwards. The gla.s.s-paper file before referred to will now come into service; it should be made of a nicely-squared plate of wood about six inches in length by about two and a half inches in width, with about one third of an inch in depth. We may call one side the front, the other the back; the edges of the former should be rounded down to a semi-circular form. All we have to do in making this useful file complete is to lap a piece of gla.s.spaper of the degree of grain required round it, nearly meeting at what may be called the back for the time being. The surface with the rounded edging, or, as we have called it, the front, covered with rather strong grained paper, will be worked to make an even course all along the shaft, guarding all the while against working too much at one spot. The paper file may, after a sufficient working along the surface, be exchanged for one or two degrees finer, rubbing it in the same manner. For good mathematically even work, the graft should be turned round frequently, so that the light may throw up any little inaccuracies that may occur and which require individual attention. Here it may be remarked that if possible all repairing would be best done in an apartment that has as little reflected light as possible. The reason for this will soon be apparent when the fining down or polishing stage is commenced. One window, and that not too large, will be found advantageous. Little irregularities, however trifling, are best seen under such a light. Much rough work may very possibly be the consequence of badly-arranged light rather than inability or indifference of the workman. Repairs executed under unfavourable circ.u.mstances as above will often look very well until turned about in fresh and different lights, as they are sure to be, and then the faultiness becomes a surprise to the executant.

The gla.s.spaper filing must be continued for some time and with several finer degrees until the surface appears perfectly even and seemingly quite finished, but the stages are not yet complete.