CHAPTER VI.

THE CHAIN AND ITS FUNCTIONS.

There are few, if any, parts of the modern bicycle that have played a more important rôle in its development, than has the chain, and yet it is safe to say that there is no part of the vehicle to which the average rider pays less attention, save to occasionally clean it of its accumulated impediments, or which he understands so little.

Every rider, of course, understands how important is the office of the chain in the propulsion of his wheel—that without it his machine is an utterly useless structure of metal, wood and rubber. As to its parts, however, and their relation to one another, he is oftener than not carelessly indifferent. While as to the mechanical skill and genius that has overcome, one by one, the past difficulties of chain and sprocket propulsion, as applied to the bicycle, bringing it in the end, to its present state of perfection, he is wholly uninformed. Many riders have been inconvenienced and annoyed to the extent of exasperation, upon discovering that “something was wrong” about their wheels. Just what, they have been utterly at a loss to tell or understand, but the fact has remained that “something was wrong,” and so, cutting their rides short, they have despatched their wheel forthwith to the repair shop. Had they known, as the repair man knew, that it was their own lack of familiarity and consequent sense of appreciation of that apparently simple, yet sensitive part of their machines—the chain, to which their misfortunes were due, how great would have been their astonishment.

It is with a thorough appreciation of how large a percentage of the wheelman’s misfortunes are chargeable to a lack of knowledge of chain construction and action, that the writer has deemed the subject one well worthy of special treatment in these columns. That many readers will admit, after perusal, that however well they may have understood their wheels in other and less important parts, they still had much to learn of its most vital and intricate parts, is altogether likely.

A study of cycle chain construction will show the regulation chain to be, simply speaking, an endless belt provided with holes which engage projections on a form of pulley called a sprocket. It is composed of blocks alternating with and joined by a pair of links or side-plates; the blocks drop down into the spaces between the teeth on the sprocket, and those teeth come up through the spaces or openings between each two side links, these links of course holding the whole together by pins through their ends.

The “pitch” of a sprocket, as of any toothed wheel, means usually the number of teeth cut upon it for each inch of its diameter. The “pitch-line” is circumferential, though not at the extreme ends of the teeth; it is the line where the teeth of two engaging gear wheels come together, or the line passing through the contact or acting surfaces of the teeth. As a chain lies on the sprockets, this pitch line passes almost exactly through the centre of the teeth, and the rivets of the chain.

To speak of a chain as “one-quarter-inch” or as a “three-sixteenths chain” means that such is its measure in width between the plates of the links. This is also the thickness of the sprocket, barring a very slight difference to prevent too tight a fit. To speak of a chain as having an “inch pitch” (which is the regular standard in this country) means that the distance between the centres of the spaces through which the sprocket teeth come (as above stated) is one inch, and of course the same measurement applies to the sprocket; the spaces on that, measured between the centres of two adjacent teeth, must be an inch. It is plain that sprocket and chain must correspond in order to work properly. A chain of a half-inch pitch would not fit a sprocket of one-inch pitch, or vice versa. If the chain were made just a little too “long,” it might go part way around the sprocket, but a disagreement would soon be found. It is charged against the chain, and correctly, that use (helped by dirt under the condition of being uncovered) wears chain and sprocket both, so that they gradually cease to match together, as at first. When this occurs, the chain is said to be “out of pitch.” On the other hand, a chain will work a long time and very well after it has considerably lost its first exactness of fit, whereas gears which have worn grind and complain dismally.

To arrange the lines of gear teeth, either straight or by various gentle curves, so that when the teeth are in operation they will close together and then separate with a rolling motion, with no slipping or grinding, with no friction, has been a mechanical problem for a hundred years. This has not been accomplished on the bevel geared chainless bicycle, and it can never be fully accomplished anywhere. Press the palms of your hands together firmly, then slide one hand off the other while so pressed—that is rubbing friction; now lay the backs of your hands together, pressing as before, and roll them away from each other until they part at the ends of the fingers. That is rolling friction, and if we could only manage to make gears and other contacting surfaces in machinery meet and part company exactly thus, we could avoid friction almost altogether.

OPERATION AND EVOLUTION OF THE CHAIN.

In considering the chain most people forget that although made up of many pieces of metal only a few are in action at a time. Only the upper half is in tension (the action is, of course, reversed in back-pedalling), and if the chain is opened and allowed to drop down it will for the moment act just the same. It is full of joints, but few are bending at any instant. As the chain runs upon the sprocket, its joints bend to conform to the circle, and they similarly bend back to an approximately straight line when leaving it. On the lower side, the joints bend easily; on the upper, they do so under tension. Press your thumb on the palm of the other hand, and, while pressing hard, draw it off; this gives some idea of the rubbing friction when the chain block leaves the tooth against which it is pulling. There is also some rub on the tooth where the chain is coming on the sprocket; and unless it is avoided by devices to be presently described, there is a rubbing between the tooth and the ends of the bending links, as well as within the joints themselves when they bend under pull. The effect of this friction is shown in the wear which comes on portions of the teeth; it also shows by flat places worn on the chain blocks, and the wear within the joints causes what is called “stretch,” the chain appearing to have grown longer. In a very slight degree there is a yielding between the parts which is called “set,” parts which are already in contact being pressed into still closer contact; this “set” supplies the trifle of elasticity, already mentioned, which tends to save the chain from fracture under heavy stress.

Chain and sprocket act on each other much as the teeth of gears act, and in effect they are a peculiar form of gears, for if you can imagine one of a pair of gear wheels flexible and flattened out like a chain, and thus running, it is evident that this action is really that of gearing. Chains were used on the tricycle before they were required for bicycles, and as long ago as 1881 there was a substitute attempted which was described thus: “The Queen driving bands are made very thin and neat, of a compound of silk and other strong substances, and are substituted for chains to save both weight, noise, and appearance.” The early chains were heavy and wide, at least ⅝-inch, and crudely made. The Ewart, as used on the Columbia Veloce ten years ago, was ingenious and simple; block and side-link were one, there was neither special joint nor rivet, and the chain could be opened at any point by turning it (see cut on page 61) and sliding to one side. Width of chain and thickness of sprocket gradually lessened; a few years ago, ¼ was the standard, but now it has settled to 3/16, even on tandems, and on racing wheels a ⅛ chain has been used in a few instances. The “B” chain has almost displaced the “8.”

Quality of steel used, accuracy of pitch and fineness of fit and finish have steadily improved, and were never at so high a standard as in 1898; accuracy of cutting and scientific shaping of the sprocket teeth have also been constantly studied and show greater advance than ever, so that, as a result, the chains on this year’s product run with a smoothness and “sweetness” not before attained. The chainless movement has naturally contributed to this advance, which is a substantial fact. Quality has improved while cost of production and market price have declined, and the high-grade chain of 1898 may without extravagance of language be called “beautiful.” For instance, in a specimen before us the blocks are nickel steel, straw color, and the links are of bright tool steel; the inner edges of the links are chamfered or beveled to lessen the chance of the chain’s ever “mounting the sprocket” if it is run when too slack, and the ends of the pins are so perfectly headed that the operation has left no trace. This finely finished specimen happens to be from the Lefevre Arms Company of Syracuse, but like praise can be given to the best product of several other makers. As to strength, chains used to be made with a breaking strain as high as 1,800 pounds; we suppose the average with the narrower and lighter product of today is about 1,000 pounds, which is far beyond any driving strain it can receive.

ATTEMPTS TO DEAL WITH CHAIN FRICTION.

The friction of the chain is of three sorts and at three places. First is the “block” friction, where and when a few blocks at a time enter and leave contact with the sprockets on the upper side, the action on the lower side (except in back-pedalling) being so free that it need not be taken into account; the second is the “pin” friction, made by the side links as they turn on the rivets; the third is where and when the ends of the links rub on the sprockets while bending.

There have been many attempts to turn these rubbing frictions into rolling movements. Only a few months ago application was filed for an English patent on putting balls into the chain joints; but the great number of joints and the small size of the parts make this plainly impracticable. A far better and really practical thing is the Morse roller-joint chain, made in Trumansburg, N. Y., and now in use on several makes, among them the Sterling; it would undoubtedly make its way faster into use except that the parts have to be a little larger, and therefore the pitch a little more than the regular inch, and so the sprockets must be cut specially; sprockets of inch pitch can, however, be recut to fit. As the illustration shows, the principle of this joint is the same as that of scales—the knife-edge bearing. The pin with the two edges is fast to the side link; the pin with one edge is dropped in loosely and the two rock on each other instead of rubbing, producing no wear and so not needing lubrication. The maker claims a frictional loss by his chain of less than one per cent. of the power developed; there is always some loss, it should be observed, and so the advertisement of the Eadie roller-chain, that “it transmits practically 100 per cent. of the force applied,” is somewhat too enthusiastically worded.

THE BROWN ROLLER-SPROCKET.

(See page 58.)

While giving due credit to the Morse chain, we must point out that it attempts to deal with only the second of the sources of friction above stated. The Brown roller-sprocket apparently attempts to deal with all three, involving an action unlike any other. It has a rim with a double flange, in which are inserted hardened steel rollers three-quarters of an inch in diameter, running on hardened steel bushings, which in turn are free to revolve on hard steel rivets. The chain is 3/16 and of 1¼ inches pitch; it is reversible and the side links are longer than the blocks, which in action ride over the rollers, reaching from one roller to another without touching the rim of the sprocket. Instead of the block rubbing on the tooth as it leaves the sprocket, it turns the roller and rolls off; thus, if the stress of use develops no other action of the parts than is claimed, the only rubbing friction is at the axes of the rollers, where the motion is comparatively slight. A drawback is that the sprockets must be very large in order to get a goodly number of rollers in the rear one, and the same difficulty of being special in both sprocket and chain, which retards some other devices in the market probably affects this one.

VARIOUS ROLLER-CHAINS.

The twin roller has entirely displaced the single. The value of the roller depends upon the difference in diameter between the roller itself and the axis on which it turns, the theory being that although there is a rubbing friction on the axis, the motion there is so slight as to be insignificant as respects wear. The smaller the roller the less this theory applies and the less the practical effect in reducing friction. Rollers in a chain are necessarily small; yet when the roller pulls off the sprocket tooth under pressure it is free to turn, and so there must be some lessening of friction—at least, the rollers cannot wear into flat spots as the blocks usually do. The twin-roller was hailed with satisfaction in England, a year ago, the chief mechanical authority in the trade press saying that “after using it for weeks in all sorts of weather we are firmly convinced that it is the chain of the future; in a gear-case it runs as smooth as oil, and even when unprotected and smothered in mud, dirt and grit seem to have little effect on it.” Since then some doubtful or dissentient opinions have been expressed, perhaps because some makers cut up this chain into a shorter pitch, and therefore get it slighter and more exposed to clogging. Without having practical experience of the twin-roller as yet, we strongly incline to agree with the opinion of it just quoted, and all theory is certainly in its favor. It has been regularly used on the Keating during 1897, and seems to be coming on.

The Thames chain, which is called a “roller block” instead of a twin-roller, has the peculiarity of a fixed cross-bar (very poorly shown in the cut) between each pair of rollers constituting a “block.” Thus the “centre block” is claimed to be rigid and the rollers to be kept more free to work; it seems to us, however, that the roller is slightly too small to be in the best proportion to the side plates.

THE LINK FRICTION ON SPROCKET AND PIN.

As to the third of the three chain frictions above described—that of the ends of the links on the sprocket as they bend into or out of the straight line—a serious practical question is involved. Plainly, as the chain is pulled hard toward the centre of the sprocket, it must come to a firm rest on something; what shall that something be? The ends of two adjacent blocks may come to a stop on the sloping sides of the tooth or either the ends of the blocks or the ends of the side-plates (or possibly both) may rest on the space on the sprocket between the teeth; or the side-plates may have a resting place outside the teeth. There has been a flange on the sprocket, just at the base of the teeth, sometimes on both sides and sometimes on only one; this flange, called a “shroud” in England, has been quite a subject of discussion there, as to its proper purpose, and even whether it should be on the sprocket at all. All agree that it is useful in stiffening the sprocket laterally, and some, including some of the best chain makers, argue that it ought to be placed below the reach of the chain, for if the chain touches it and wears it away, the chain will sink below its correct pitch line and cause trouble. Others claim that the chief object of making a shroud was to give the chain as much bearing surface as possible. Practice is not uniform in either country. Sometimes the shroud has helped support the chain; sometimes it has been kept out of reach of the chain, and sometimes it has been cut away where the ends of the side-plates come.

The pins are, of course, fixed in the links, but form a bearing within the ends of the blocks. To have these pins hard and yet be able to “upset” their ends to make a “head” has been a matter of difficulty. Some of the best English makers avoided this by using a soft pin and putting a hard bushing of pen steel over it to make a bearing. As the links do not need to be hard, the makers of the Cleveland use a hard pin with a groove at the end, and force the end of the side-plates into this groove to hold the pin in place. Other chain makers have contrived methods of getting the pin hard and still having its ends capable of being headed over. The Myers Detachable, made by the Bridgeport Chain Company of Bridgeport, Conn., and the Baldwin Detachable, made by the Baldwin Chain Company of Worcester, Mass., avoid the difficulty by hardening the entire pin and slotting the link plates, as shown in the illustration. Any broken piece can thus be replaced, or the chain can be made longer or shorter at will, without needing any tools. The Baldwin pattern is reversible, and the makers publish a certificate of one of their chains, which has a record of 29,573 miles; of this, 13,771 miles were done without any attention or repairs being required, the remaining mileage requiring replacement of a broken part but twice.

We find in one of the British trade journals a mention of a “spring chain,” but there is neither cut nor detailed description. The maker claims that by inserting a dozen or so of his spring links in any suitable chain “it will be made to run as easily and smoothly as a leather driving band, and that it may be adjusted so tightly as to practically do away with all slackening on top, so that every ounce of driving pressure applied to the pedals will be reproduced on the rear sprocket wheel, thus getting rid of all backlash and consequent friction and waste of power, even when ridden over the roughest roads and by the most inexperienced pedallers.” The editor thinks it impossible, without some dynamometer test, to say whether there is any gain in driving ease, but after having one of the chains in use for a good many months he can bear out the claim of smooth running, and has found that it can be run on a tighter adjustment than the usual chain; so “it certainly seems to be satisfactory.”

The circular chain is another peculiar English device, and is pronounced by its maker to be the best and easiest running, wear-resisting and cleanest he has ever tried. His claim is: “The circumferential speed of the block chain is in excess of the corresponding speed of its chain wheel teeth, hence the contradictory friction between tooth and block. The circumferential speed of this chain and its wheel are similar; entirely does away the frictional contact between the teeth of the chain wheel and the chain blocks.”

His first statement cannot possibly be correct so long as the pitch line of the sprocket and of the chain correspond, as is the case with any reasonably good fit. His chain consists of simple and uniform links, turned from the solid and joined by rivets. It suggests the old chain of the chain pump, and, of course, requires a peculiar sprocket.

The Tacagni standard pivot or rivetless is a recently offered English article. It is light, weighing 7¾ ounces, against the usual 14½. Less friction and greater strength are also claimed for it, the maker offering the report of a testing firm that the elastic limit of the sample used was 900 pounds and its breaking stress was 1440. Of course, a special sprocket grooved in the rim must be made for it.

THE REMINGTON CHAIN.

The chain brought out by the Remington people for their $75 model suggests the Tacagni, but is not quite like that. The illustration shows its construction. The block is done away with, since it runs in a groove on the sprocket rim; the usual link does the pulling instead, bearing on the flanges. Another style of description is to say that the construction is reversed, the link being one piece and central, being converted into a block, a space being cut out so that it does not touch the sprocket in bending, while the usual block is doubled and runs on the flanges. The same quality of steel is used throughout, and the grain all runs lengthwise. Strength, great endurance and a reduction of a fourth in friction are claimed. Use must decide the degree of improvement, but the chain certainly runs smoothly and attractively.

THE LIBERTY SPROCKET.

(See page 62.)

The Liberty makers have brought out a sprocket with a change in the form of the teeth which is so slight that it hardly shows in a cut and is not even noticed on the bicycle itself at a careless glance. The change consists in cutting down most of the teeth in height and thickness, so that only each fifth tooth acts in the driving, the intermediate teeth serving only as guides to keep the chain in track.

Concerning the new sprocket, the makers say: “The old method of having each and every sprocket tooth engage the chain has been abandoned by us, the friction occasioned by so much contact being unnecessary, and the wear and strain on the chain intensified. Our new sprocket has been tested under all conditions with the most satisfactory results. It permits a chain to run as smoothly covered with mud and dust as it does when thoroughly lubricated, and the cracking noise so prevalent when an ordinary sprocket is used on muddy roads is entirely absent. The ease of propulsion is marked (particularly noticeable in hill-climbing) and enables the rider to attain speed instantly and with the highest gear. With this improved sprocket the rider can use with ease a gear considerably higher than he could attempt with sprockets of the ordinary pattern.”

THE VICTOR STRAIGHT-LINE SPROCKET.

(See page 66.)

The Victor straight-line sprocket is peculiar in the shape of the teeth or in the shape of the spaces between the teeth. On the back side of each tooth on the front sprocket and the front side of each tooth on the rear sprocket a space is hollowed out, as shown in the cut. Ordinarily, each block and each pair of side-plates or links is deflected from a straight line when wrapped around the sprocket; but on this sprocket, as will be seen, each block and adjacent pair of side-plates form a straight line. The joints at E—E and F—F do not touch the sprocket, as it is cut away beneath them. As those portions of the chain are always in a straight line, no motion is produced in the joints there, the effect of the change being to greatly lessen friction, especially under trying conditions.

If an accurate measure of the pressure required on the pedal in order to overcome the varying resistances of surface, grade and wind, or what not, could be found, then the resistance in each case could be weighed and recorded in pounds. The Victor dynamometer—which is a peculiar pedal, containing a pair of springs, with a recording pencil and a moving roll of paper for making a record—does this weighing. Obviously the first effect of pressure on this box-like pedal is to depress the springs; and the wheel will not be moved at all until the springs have been depressed enough to represent the resistance. Suppose the total resistance is equal to lifting a weight of ten pounds, then press on the pedal; the springs will first yield until the equivalent of ten pounds is reached, then the pedal will move and the wheel will turn. If the resistance changes to fifteen and then to five pounds, the springs will yield more and then less, and the pencil attached will register accordingly, the result being an irregular line similar to that on the steam engineer’s “indicator card.”

If the dynamometer pedal were used on the road the irregular line on the card would show resistance fluctuations, but would not show the various times and causes of resistance encountered. So, for a test of the peculiar sprocket, a bicycle fitted with it was put on a stand and a resistance equal to seven and six-tenths pounds at the rim of the wheel was arranged. Then mud was daubed on the chain, and pressure was put on the dynamometer pedal. The height of the wave line above the straight or zero line in the diagram indicates the power required to turn the wheel. It ranged from 88 to 94 pounds, and was nearly uniform. Then another bicycle, with usual sprockets, was set on the stand, with the same resistance at the wheel rim. The same chain used before was put on (for this is a matter of sprocket only, and any usual chain both fits and answers the purpose), mud was again daubed on the chain, and the test was made. The pull required to turn each ranged from 96 to 160 pounds, and fluctuated greatly, as indicated in the other diagram. When the tests were carried further and the resistance at the tire was brought up to 11¼ pounds the ordinary sprocket clogged under the mud and could not be turned at all, while the straight-line sprocket moved about as before, the card indicating a pull of 128 to 131½ pounds applied.

At the Victor branch in Warren street, a bicycle with these sprockets rests on a stand, with a box of Jersey mud and a dish of water and a trowel underneath; anybody is free to mix the compound to suit, and to load on all the chain will carry. Then he may get on the saddle, there being an adjustable brake for the rear wheel to represent road resistance, and pedal away; or he may turn by hand. When the mudded slack of chain reaches the rear sprocket, the first effect is a crunching noise and a partial stoppage; this ceases when one revolution has been made, and directly the wheel (the brake being off) spins as freely and quietly as does another bicycle with the like sprockets which is mounted, all clean, on another stand. The mud test is actual and fair. The snapping noise which every rider knows is produced by mud on the chain, especially when the bicycle is new and the fit is at its best, comes because the mud acts as a wedge between sprocket and chain and the latter is temporarily put “out of pitch.” The surprising performance of this sprocket under the severest possible mud test can have only this explanation: that the spaces cut away allow some room for mud without jamming, and that the sprocket clears itself by throwing out the intruder. It is certainly one of the most remarkable things of the season, and seems quite independent of outside disturbance, hardly needing a case except for cleanliness and length of wear.

CHAIN BOLTS AND REPAIRS.

Of course, the ends of the chain have to be joined, and sometimes they need to be separated for removal. The customary way has been to use a screw-bolt, threaded into the link-plate on one side, and fastened with a small lock-nut. As this small nut was liable to loosen and be lost, and as there was also a possibility of the screw itself working out (in which case it might strike something as the chain moved or might drop out on the road) some securer fastening became desirable. The “Diamond” B chain now dispenses with the nut (as shown in the cut of that make of chains) using in place of it a swinging “latch” of thin steel; the head of the screw fits nearly flush into the side-link, and the latch has a place raised up in one end to fit the screw head, so that when this latch is turned down it snaps into place, preventing the screw from backing out and being itself held fast by its own elasticity. The Humber carries on its chain a similar latch, but slightly different in shape at the end, which has a hexagonal hole that fits the head of the screw. The Crescent meets the case by dispensing with the screw-bolt. As shown in the cut, the side links are slotted, and in the centre of the slot is an enlarged place through which a special pin with grooved ends can be slipped in or out by slacking the chain for the purpose.

These several devices go to further lessen the troubles with chains which are so great now, in the argument of some people, but have been so slight in practice notwithstanding.

In the very rare event of a chain’s breaking on the road, the Missing Link will be handy; it costs but a few cents, and can be carried in a vest pocket. The cut explains its use. A break is most liable to be in the block, but if a link goes the rider need not tear his hair; there are devices to meet that case, and to get another piece in is not very severe, even without their aid. Chains are “stretching, breaking,” etc., in their habits, we are told. Yet each rider may consider the chance of the trick’s being played on him nearly the same as of lightning’s striking him, and if he will only take a little care of his chain, he can count himself insured.

CHAIN ADJUSTMENTS.

Some form of adjuster will always be necessary to adjust the chain on a chain driven bicycle. On the early models of the safety type of bicycles made in this country the adjustment was produced by a swinging crank bracket. The crank bracket was not an integral part of the frame, but was bolted to it and was held in position by a set screw and lock-nut. Somewhat later an improved form, which by the usual form of reversion has now come into use again, consisted in making the crank bracket an integral part of the frame and fitting an eccentric adjustment inside of it. The Remington Company varied this somewhat by making the rear forks a detachable part of the frame and having them bolted through and locked by a threaded lock-nut and bolt at the crank hanger, and they thus produced their adjustment by shifting the rear forks out backward or drawing them forward. Since that time the makers of the Remington have always used the rear fork-end adjustment, but this season they have a new feature. The crank-hanger ball pocket is eccentric and turns in the bracket either forward or backward when the set bolts are loosened. The whole arrangement is a very simple one and prevents the liability of the rear wheel getting out of alignment.

The makers of the Iroquois also use a 3-inch eccentric hanger. The rear wheel is always centred by this method, and is provided with two sprockets, so as readily to allow a change of gear.

On the Defender is shown an eccentric crank-hanger, on which neither the wheel, nuts or bearings are disturbed to make the adjustment.

The Shirk bicycles have a new rear fork and chain adjustment, the advantage of which is that the rear wheel can be removed without disconnecting the chain. The sides of the rear fork ends are machined with teeth, which fit into the teeth of the washer, and by simply unscrewing nut and withdrawing the axle bolt the wheel drops out of frame. Absolute equality of adjustment on both sides is obtained, as the wrench is only used to loosen the axle nut, and as the outward opening the rear fork ends is done away with, strength and rigidity is thus added to this end of the frame.

The makers of the Northampton made a new chain adjuster consisting of a small round steel plate on the outside of the rear forks, with scroll cut on the inside which follows steel lug on the forks, making it easy to adjust chain to any tension and set the wheel true in the frame rapidly.

The chain adjuster used on the Globe is of very neat and simple construction. A threaded adjuster, having an open hook end is pivoted to the upper part of the rear fork end, and is operated as follows: Loosen the axle nuts and turn the thumb screw either way, as the case may be, until the chain has the right tension and then tighten the axle nuts again. To take the rear wheel out, loosen the axle nuts and swing the hooks off the axle. To replace the wheel slip the hooks back over the axle, tighten the nuts and the whole adjustment is complete.

On the Relay is shown a patent chain-adjusting device which enables the rider by simply loosening the nuts on either side of the rear wheel to remove the rear wheel without taking the chain apart. The fork ends are of cold rolled steel, corrugated, with the washer corrugated to correspond, allowing accurate adjustment of the chain.

On the model 4 Humber is shown a rear fork chain adjuster, which is similar in construction to the chain adjusters in use on the Humbers made in England. The rear fork ends instead of being carried horizontal as before now slant upward at an oblique angle, and the backstays instead of being brazed to the rear forks as heretofore are separate and are carried backward or forward, as the case may be, with the rear axle to tighten or loosen the chain, the object of this change in construction being to cause the backstays to help carry with the rear forks the weight of the rider on the axle.

The Wolff-American patent eccentric chain adjuster is almost too well known to need describing. Still it is such a radical departure, and withal such a good one, that it will bear describing here again. A square groove or spline is cut on the sides of the rear axle, running about an inch from the end. A pair of eccentric disks, having a tongue or key to fit this groove, are slipped on the axle, thus becoming, as it were, a part of the rear axle. They are then placed and held in the frame by semi-circular braces, which are a part of the frame. The chain is adjusted from one side, the eccentrics acting together. By loosening one nut on each side the eccentrics are free to move either way. This completes the operation, and, it is needless to say, one need not worry about getting the rear wheel out of line or readjusting the bearings, because with this eccentric adjuster neither is disturbed. They use the same method of adjustment on the rear wheel of their tandems, but the front chain on the tandems is adjusted with an eccentric at the front crank-hanger, same as most of the other makers use in tandem construction. Nearly all the makers who make tandems adjust their rear wheel, however, with their regular form of chain adjuster as used on their singles, a variation of this, however, being to adjust both chains at the crank-hanger brackets with an eccentric adjustment.

Another variation in chain adjustments on tandems consists of bolting the crank bracket to the frame so that by moving the crank bracket forward or backward the chain can be adjusted to the proper tension. The makers of the juvenile “Elfin” not only use this form of construction on their juvenile tandems, but also on their single models, and have in addition to that a method of reversing the bracket, so that it can be either bolted on top or underneath the rear forks which permits an adjustability of two inches between the seat posts and pedals, by which an Elfin may be made to last a growing child for several seasons.