CHAPTER V.

ON TUMBLER, OR LEVER LOCKS.

Security being the primary object in all locks, any considerations as to mechanical ingenuity and graceful decoration give place to those which relate to safety. A spring lock may be ingenious and even beautiful in its construction, but an imitative key will easily open it. Hence arose the invention of wheels or wards; and as wards failed in trustworthiness, they in their turn yielded to something better. We have already explained how the insecurity of mere warded locks arises; and we shall have something more to say on the subject in a future chapter. It is sufficient here to remark, that wards, springs, screws, alarums, wheel-work, escutcheons,--all, however useful for particular purposes, are wanting in the degree of surety which we require in a lock. Hence the invention of _tumblers_, _levers_, or _latches_, which fall into the bolt and prevent it from being shot until they have been raised or released by the action of the key. We have been unable to ascertain at what time, or in what country, or by whom, tumbler-locks were invented. The invention has been claimed by or for persons subsequently to the year 1767, when the celebrated French treatise (_Art du Serrurier_) already referred to was published; and yet this treatise contains numerous examples of simple tumbler locks of ingenious construction, as will presently be shewn.

One of the most elementary forms of tumbler-lock is shewn in fig. 17. In this case the bolt, instead of having two notches in the bottom edge, like those in the back-spring lock, fig. 6, has two square notches or slots in the upper edge; and as the key acts upon the bolt, these notches must of course share in whatever movements the bolt is subjected to. Behind the bolt is a kind of latch or tumbler (the lower part of which is shewn by dotted lines), with a stump or projecting piece of metal at _a_; the tumbler moves freely on a pivot at the other end, and is made to rise through a small arc whenever the key acts upon the bolt. When the bolt is wholly shot, the stump falls into one notch and prevents the motion of the bolt; when wholly unshot or withdrawn, the stump falls into the other notch, and equally prevents the motion of the bolt. It is not, therefore, until the key, by elevating the tumbler, has raised the stump out of the notch, that the bolt has freedom of movement. If the shape of the key does not enable its web to effect this elevation to a sufficient degree, the bolt remains immovable; and to this extent a certain additional security is obtained by making the shape of the key significant as well as the wards.

The tumbler-principle, as we have said, is difficult to trace to its origin on account of the various aspects which it presents; but the great French treatise proves that the locksmiths of France were familiar with tumbler-locks a century ago. The plates of that work represent the details of numerous locks, on the upper edge of the bolts of which were notches called _encoches_, as at _o k_ fig. 18; into these notches sank a small iron stud or stump called the _arrêt du pêne_, or bolt-stop, shewn in fig. 19, attached to the upper portion of the _gâchette_ or tumbler, which, for the sake of economy of metal, is made in the form of a triangular spring in front of the bolt _k i_; and not until the key, by its circular action, had raised this stud out of one or other of the notches, could the bolt move to the right or left. The stud was generally fixed to a spring which forced it down again into the notch as soon as the action of the key had ceased. Sometimes, however, the stud was fixed to the bolt, and the notches were in a separate tumbler or _gâchette_ (see E E, fig. 21); and in other instances, again, the stump was fixed to the case of the lock and caught into notches in the bolt. It will be seen, when we come to treat of tumbler-locks of later date, that there was much in these early locks to point out the way. Fig. 19, copied from the French work, represents a lock of the box or casket kind. Two staples, fixed into the cover, fall into two cavities or receptacles at C _d_; and a short bolt in each receptacle catches into each staple, one near _g_ and one near _h_. The small bolt _q_ is attached to the upper extremity of the lever _q r s_, fig. 19, and shewn separately in fig. 20; and by the pressure of a spring _a_ (fig. 19) upon this lever, the bolt _q_ is kept locked in the staple. The vertical portion of this spring presses at its lower end on another spring _p_ (fig. 19) of singular curvature; and attached to the horizontal part of this second spring is the stud, which falls into a notch in the top of the bolt. The action of these parts, then, is as follows: when the key is placed upon the key-pin at Z, and turned round in the direction in which the hands of a watch move, the bitt presses against the tail _s_ of the lever, moves it upon its centre Z, fig. 19, _v_, fig. 20, to the left, and consequently moves the upper part _q_ to the right, drawing it out of the receptacle and liberating the staple within C. Thus it will be seen that the lever _q r s_, held in one position by the spring _a_, forms in itself a simple kind of spring catch-lock, and was, in fact, formerly used as such, without any other appendages except the staple in the lever, into which the catch _q_ fitted on shutting down the lid. So also we may regard the other portion, fig. 18, or _k i p h_ (fig. 19), as forming a separate lock; for the key after having passed S comes in contact with the triangular spring, which it raises thereby, lifting the stud out of the bolt, and exerting pressure against the barbs of the bolt _n_. Fig. 18 shoots the bolt _k_, and also the short bolt _l_, which passes through the staple in the cavity _d_, fig. 19.

The lock represented in the four following figures is also from M. de Réaumur’s chapter on locks in the work referred to. In this lock the tumbler-principle is carried out in a very elaborate manner, for not only is the stump or stud H (fig. 23) attached to a very strong spring (best shewn at H, fig. 22), which holds it with considerable force in one of the three notches of the principal bolt R S (fig. 24); but there is also a second set of notches E E in the _gâchette_ G O (fig. 21), and a pin attached to one of the plates of the lock fits into one of these notches, thereby preventing the bolt from being moved until the _gâchette_ is lowered by the revolution of the key; so that in attempting to pick this lock, not only must the spring H be raised so as to release the stud from the notches of the great bolt, but the _gâchette_ must be lowered to disengage the fixed pin from the notches. There is yet a third source of security. Attached to the large bolt are short projecting pins F (fig. 21), against which an arm or detent, G F, of the _gâchette_ projects, thus preventing the bolt from being shot back by any pressure applied to its extremity S.

There are a few details relating to this remarkable lock, which may as well be introduced here in order to complete the description. The principal bolt can be shot twice, or be _double-locked_; hence it is furnished with three barbs for the key to act against, and with three notches for the spring-stud. The lower bolt I K can be shot by the horizontal pressure of the button P (figs. 22, 23), which is situated on the inner side of the door to which this lock is attached, so that a person inside the room can secure the door against any one on the outside who is not furnished with the proper key, for it must be remarked that the small bolt as well as the large one is acted on by the key. Now supposing the small bolt to be shot or locked, it is kept so by the pressure of the coiled spring Q (figs. 21, 22). But this small bolt is connected with the large one by means of the bent lever O N M (figs. 21, 24), which turns on a pin N attached to the main bolt. Now, when both bolts are either fully shot or unshot, the arm O N lies flat against and parallel with the main bolt; but when the large bolt is unshot and the small one not moved, the arms O N, N M, fall into an inclined position, and the arm O N passing a little below the main bolt comes within the range of the web of the key, which in its revolution causes the bent lever to move upon its centre N, thereby restoring O N to its horizontal position, and at the same time causing the arm N M to move from right to left, or in the direction for unshooting the small bolt; the end of this arm thus catches into a mortise V (figs. 21, 24) in the small bolt, and immediately unlocks it.

But to return to the subject of tumbler-locks. About the year 1778, Mr. Barron introduced that species of double-action (as it may perhaps be termed) which so greatly increases the security of the simple tumbler, fig. 17. In the tumbler-locks previously made, if the tumbler were raised sufficiently high, the lock could be opened: there was no such possibility as raising it _too_ high; but Mr. Barron, by his invention, patented 31st October, 1778, rendered it absolutely necessary that a limit should be put to the height to which the tumbler should be raised, by rendering the bolt equally immovable whether the tumbler were too much or too little raised. Another important improvement was the introduction of two tumblers instead of one. The bolt has in its middle a slot or gating notched on both edges, the notches being fitted for the reception of studs fixed to the tumblers. Supposing the studs or stumps of the tumblers to be resting in the lower notches, they require to be elevated to the general level of the gating before the bolt can be moved; whereas, on the other hand, if the tumblers were raised ever so little too high, the studs will enter the upper notches, and prevent the shooting of the bolt. The lower edge, or belly, of each tumbler is acted on by the steps of the key during its circular movement; the leverage of the key being so exactly adjusted as to raise the tumbler to the desired height and no further. The tumblers are made unequally wide, so that steps or inequalities in the bit of the key are requisite to lift them both to the proper height. There are thus two improvements introduced: there are two tumblers instead of one, and each tumbler has a double instead of a single action.

This ingenious and very useful lock is represented, so far as regards its governing principle, in fig. 25. The bolt is here seen to have a peculiar slot or hole cut in it, consisting of a narrow horizontal passage or gating, with three notches above it and three below it. These double notches might be available even for one tumbler only; but Barron used two or more for the sake of additional security. In fig. 25 there are two tumblers shewn, expressed by dotted lines; both are hinged to one pivot, both are raised by the same action of the key, but the stump on the one tumbler does not coincide in position with that on the other. It will be seen that if the studs of the tumblers rested in the lower notches, they would require to be elevated to the level of the gating before the bolt could be moved; while, on the other hand, if lifted too high, the stumps would be caught in the upper notches, and would equally prevent the passage of the bolt, The tumblers are unequally wide; and the bitt of the key is stepped or notched in a corresponding way, that there may be one step fitted to act upon each tumbler. Mr. Barron also adopted the reverse arrangement of having the stump on the bolt, and the openings in the tumblers; so that the principle of his patent may be concisely expressed as being “an arrangement to allow a stump on the tumbler to pass through an opening in the bolt, or a stump on the bolt to pass through an opening in the tumbler.”

A very elaborate tumbler-lock, patented 23d February, 1790, by Mr. Rowntree, contrasts remarkably with the simplicity of Barron’s lock. Mr. Rowntree’s lock consisted of tumblers combined with revolving discs or wheels. Its mechanism may be understood from the following description and engravings. The same letters refer to the same parts in the several figures.

A A is the plate which encloses the whole mechanism of the lock, and fastens it to the door; B B is the bolt, guided in its motion by sliding under the bridges C D; E E are pillars which support a plate covering the works; F are the circular wards surrounding the centre or key-pin; and _a_ shews the position of the key, which, in turning round, acts in a notch _r_ in the bolt, and propels it; G, the tumbler, is a plate situated beneath the bolt, and moving on a centre-pin at _d_; it has a catch or stump _e_ projecting upwards, which enters the notches _s_ or _g_ in the bolt, and thereby retains the latter for backward or forward motion, as the case may be; H is a spring which presses the tumbler forward. The key _a_, in turning round, acts first against the part _c c_ of the tumbler, and raises it so as to remove the stump from the notches; it can then enter the notch _r_ in the bolt, and move it. So far there is no particular security; but Mr. Rowntree sought to obtain it by the following means. There is a piece of metal _h_ fixed to the lower side of the tumbler, called the _pin_; when the tumbler is caught in either notch of the bolt, the pin applies itself to a cluster of small wheels I, fitted on one centre-pin beneath the tumbler; the edges of these wheels stop the pin, and prevent the tumbler from being raised. But each wheel has a notch cut in its circumference I; and it is only when the wheels are so placed that all their notches lie in a right line, that the pin can enter this compound notch and allow the tumbler to rise. The wheels must therefore be all adjusted to position; and this is effected by a number of levers K centred on one pin at _k_; at the opposite end each lever has a tooth _m_ entering a notch in the wheel belonging to it; so that when any lever is pressed outward, it turns its wheel round. Now this pressure of the levers is brought about by a spring _n_ applied to each; and when so pressed, the levers rest against a pin _o_ fixed in the plate. The key is so cut as to determine the extent to which the levers shall act upon the wheels. The key first operates from the curved part _p p_ of the levers K, and raising them, turns all the wheels I at once into the proper positions; in turning further round, it then operates on the part _c c_ of the tumbler, causing the latter to rise and to release the bolt; and in turning still further round, it (the key) seizes the notch _r_ of the bolt, and shoots it. The key is cut into steps of different lengths, as shewn at V V; each step operates on its respective lever K in a different degree from the others; the notch at _s_ acts upon the tumbler, and the plain part _t_ moves the bolt.

We now proceed to notice the modern tumbler-lock. This was arranged by Bird, whose patent, bearing date 29th October, 1790, was for a series of four double-acting tumblers, differing in no respect from those patented by Barron, and closely resembling those in use at the present time in the best tumbler-locks. We will describe the modern tumbler-lock more particularly when we have gone through a few historical details on the subject.

Messrs. Mitchell and Lawton obtained a patent bearing date 7th March, 1815, for a lock in which were combined with the bolt and double-acting tumblers, a series of movable wards, and a revolving curtain for closing the key-hole. The action of the wards was peculiar. On introducing any key or instrument, and passing it round, a number of movable wards or pieces were thrown out so as to prevent the key from being turned back or withdrawn. It was necessary therefore to pass round the key so as to unlock the lock, and if that were not possible, as in the case of a false key being used, it was held permanently, and could only be released by destroying the lock, When the bolt was once shot, the wards were carried up so as to leave a clear passage for the key. This lock does not appear ever to have come into use, on account of the violence required in case a wrong key should be used either by accident or design.

The detention of a wrong key in this lock appears to have suggested the contrivance of a _detector_. This was first made by Ruxton, whose patent is dated 14th May, 1816. His detectors were of various kinds, the object of each kind being to give information to the owner in case any one of the tumblers should be overlifted in an attempt to pick the lock, which fact would be discovered on the next application of the true key. This is precisely the object of the detector in tumbler-locks at the present day, and Ruxton accomplished it by somewhat similar means. He also had a contrivance for holding a false key, as in Mitchell and Lawton’s lock; and he recommended this form of detector in the following words: “It is true that in this case the lock will have to be destroyed in order to open the door: the result is frightful; but we think the more terrible the result, the less likely would any one be to tamper with it.”

We now come to Chubb’s lock, patented 3d February, 1818, which consisted of double-acting tumblers and a peculiar kind of detector. This lock has been made the subject of various patents obtained in the years 1824, 1833, 1846, and 1847. This lock[4] consists of six separate and distinct double-acting tumblers, all of which must be raised to a particular height, neither more nor less, in order that the bolt may pass. It also comprises a _detector_, by which, should any one of the tumblers be lifted too high in an attempt to pick or open the lock by a false key, it would be immediately detected on the next application of the proper key. The tumblers are flat pieces of iron or steel, with the plane of the surface vertical, and pivoted at one end; and the following is the mode in which the key, the tumblers, and the bolt, are brought into mutual action.

[4] The lock about to be described is the latest and most complete form of Chubb lock up to the date of the Great Exhibition. The various additions and alterations which have been made in the lock since that date will be noticed in a subsequent chapter.

The bolt shoots in and out of the lock in the usual way. It has a square stud or stump riveted on one surface; and it is to furnish obstructions to the passage of this stud that the tumblers are provided. All the six tumblers are pivoted to one pin at the end, giving to each of them a small leverage, each independent of the others. There are six springs which press these tumblers downwards, one to each tumbler. There is a longitudinal slot or gating in each tumbler, large enough to receive the stud of the bolt; and unless all the six slots (supposing there to be six tumblers) coincide in height or position, the stud will not have a clear passage for moving to and fro. Now the slots are purposely made nearer the upper edge in some of the tumblers than in others, all the six being different in this respect; so that if they are all lifted _equally_, the slots do not coincide, and the bolt and its stud will not pass. The tumblers must then be raised _unequally_, those to be most raised which have the slot nearest to the lower edge. To effect this, the bit of the key is cut into six steps or inequalities, each to act upon one particular tumbler, and each cut or stepped to the exact depth which will suffice for the proper raising of the tumbler. The key is inserted in the keyhole, and is turned; the six steps raise the six tumblers all to the proper height, to leave a clear passage along the slots; and the extreme end of the key then acts upon the bolt itself, and shoots it. To unlock it again, the same or a duplicate key must be used; for if another key be employed, differing by ever so little from the proper one, some one or more of the tumblers will be lifted either a little too much or not quite enough; and in either case the stud of the bolt will catch above or below the slot, instead of having a clear line of movement along the slot itself. After both locking and unlocking, the springs force the tumblers down as far as they can go, burying the stud in the recesses above the slot; so that the tumblers must be raised by the key both for locking and unlocking.

The doctrine of chances has wide play in determining the relative position of the six tumblers. In Mr. Chubb’s essay this part of the subject is treated in the following way: “The number of changes which may be effected on the keys of a three-inch drawer-lock is 1 × 2 × 3 × 4 × 5 × 6 = 720, the number of different combinations which may be made on the six steps of unequal lengths (on a six-tumbler lock), without altering the length of either step. The height of the shortest step is, however, capable of being reduced 20 times; and each time of being reduced, the 720 combinations may be repeated; therefore 720 × 20 = 14,400 changes. The same process, after reducing the shortest step as much as possible, may be gone through with each of the other five steps; therefore 14,400 × 6 = 86,400, which is the number of changes that can be produced on the six steps. If, however, the seventh step, which throws the bolt, be taken into account, the reduction of it only ten times would give 86,400 × 10 = 864,000, as the number of changes on locks with the keys all of one size (that is, with one key of definite size in all save the lengths of the steps). Moreover, the drill pins of the locks and the pipes of the keys may be easily made of three different sizes; and the number of changes will then be 864,000 × 3 = 2,592,000, as the whole series of changes which may be gone through with this key. In smaller keys, the steps of which are capable of being reduced only ten times, and the bolt-step only five times, the number of combinations will be 720 × 10 × 6 × 5 × 3 = 648,000. On the other hand, in larger keys, the steps of which can be reduced thirty times, and the bolt-step twenty times, the total number of combinations will be 720 × 30 × 6 × 20 × 3 = 7,776,000.”

These enormous numbers have been the cause of much of the wonderment which the six-tumbler locks have excited; and, as we shall see further on, the Bramah lock presents still more of the marvellous in respect to this ringing of the changes.

The construction and action of the Chubb lock may be further illustrated by means of an engraving, fig. 31, in which _b_ is the bolt of the lock, with a stump riveted to it marked _s_. The six tumblers are shewn perspectively, the front or anterior one being marked _t_; they all move on the centre-pin _a_, but are nevertheless perfectly distinct and separate, to allow of being elevated to different heights. At _d_ is shewn one end of a divided spring, the divisions being equal to the number of tumblers, one to each, and so bent that each spring may press upon its particular tumbler. At _e_ is the detector-spring, so placed that a projecting piece in the hindmost tumbler shall be near it; this tumbler having also fixed into it a stud or pin _p_. This being the arrangement, especially in relation to the stump _s_ and the tumblers, it follows that all the tumblers must be lifted to exact and regulated heights in order that the stump may pass through the longitudinal slits of the tumblers; unless it can do so, the bolt cannot be withdrawn. As there are gaps or notches in each tumbler both above and below the proper line of passage, and as there are no ordinary means of ascertaining when any one tumbler is lifted too high or not high enough, the safety of the lock is greatly increased by this uncertainty; especially when it is considered that this uncertainty is multiplied sixfold by the different modes in which the six tumblers are slotted. If, through the insertion of a false key, or by any other cause, any one of the tumblers be raised above its proper position, the detector spring _e_ will catch the hindmost tumbler, and retain it so as to prevent the bolt from passing; and thus, upon the next application of the true key, it will be instantly felt that some one of the tumblers has been overlifted, because the true key will not unlock it. To relieve the bolt from this temporary imprisonment, the key must be turned the reverse way, as for locking; all the tumblers will thus be brought to their proper position, and allow the stump to enter the notches _n n´_; the bevelled part of the bolt will then lift up the detector-spring, and allow the hindmost tumbler to fall down into its proper place; and all this being effected, the lock may be opened and shut in the ordinary way. The pin _p_ is so adjusted that if any one of the tumblers--front, back, or intermediate--be lifted too high, the pin will be lifted with it, and will catch into the detector-spring, thus producing the result just described.

The key is represented in fig. 32. It has six steps, besides a terminal step to act upon the bolt. The height of each step, or the distance to which it extends from the pipe of the key, depends of course on the height to which its corresponding tumbler is to be lifted; and it matters not whether the steps of the key are adjusted to the slots of the tumblers, or the slots to the steps, provided the agreement be brought about. It is simply a matter of manufacturing convenience that the key-steps are cut first and the tumbler-slots afterwards. We may here remark that _bit_, or _bitt_, is the name given, somewhat indefinitely, either to the whole flat part of a key, or to the small stepped portions of it. The flat part was formerly termed the _web_ of the key, probably from the _webbed_ appearance of the keys to complex warded locks.

After the reading of Mr. Chubb’s paper before the Institution of Civil Engineers, Mr. Owen narrated one or two circumstances connected with the early history of Chubb’s lock. A convict on board one of the prison-ships at Portsmouth dockyard, who was by profession a lock-maker, and who had been employed in London in making and repairing locks for several years, and subsequently had been notorious for picking locks, asserted that he had picked with ease one of the best of Bramah’s locks, and that he could pick Chubb’s locks with equal facility. One of the latter was secured by the seals of the late Sir George Grey, the Commissioner, and some of the principal officers of the dockyard, and given to the convict, together with files and all the tools which he stated were necessary for preparing false instruments for the purpose, as also blank keys to fit the pin of the lock. A lock exactly the same in principle was placed in his hands, that he might examine it and make himself master of its construction. If he succeeded in opening the lock, he was to receive a free pardon from the Government, and a reward of 100_l._ from Messrs. Chubb. After trying for two or three months to pick the sealed lock--during which time, by his repeated efforts, he frequently over-lifted the detector, which was as often re-adjusted for his subsequent trials--he gave up the attempt. He stated that Chubb’s were the most secure locks he had ever met with, and that it was impossible for any man to pick or to open them with false instruments.

Mr. Owen further stated, that in order to compare the merits of Bramah’s and Chubb’s locks, he had suggested a mechanical contrivance, which was applied to one of Bramah’s six-spring padlocks belonging to the Excise. It was hung upon a nail, in a vertical position, secure from lateral oscillation. A self-acting apparatus was then applied, consisting of a pipe with hexagonal grooves, and a stud or bit corresponding with the division of the lock, and secured to it by a spring. In the grooves of this pipe small slides were inserted, which pressed against the spring keys of the lock; to these slides were attached levers, acted upon by eccentrics, moved by a combination of wheels, whose teeth differed in number so as to perform the permutation required for the different depths of the spring keys, corresponding with those of the proper key to the lock. The automaton machine was set in motion by a line working over a barrel, and acted upon by a weight; and was thus left acting upon the mechanism for a considerable time. At right angles to the pipe or false key was attached a rod and weight; and when the notches in the spring keys were brought in a line with the plane of the plate or diaphragm of the lock, the rod and weight turned the false key, opened the lock, and stopped the further motion of the automaton. In that state the slides indicated the exact depth of the grooves in the proper key, and gave the form of a matrix by which to make a key similar to the original one. The automaton worked during a period varying from half an hour to three hours, according to the state of permutation of the apparatus at the moment of being applied, compared with that of the slides in the lock. We confess that it is difficult to understand the action of this automaton from Mr. Owen’s description. We imagine that the false notches would effectually prevent the operation of the instrument, and openings would be required on each slide to bring it back, so as to meet the motions of the machine.

Mr. Owen did not state whether his apparatus had been successful with one only of Bramah’s locks or with several; nor did he describe any apparatus invented with the view to the picking of Chubb’s locks. He stated, however, that in order to ascertain the effect of friction on one of these last-named locks, it was subjected to the alternate rectilinear motion of a steam-engine in Portsmouth dockyard, and was locked and unlocked upwards of 460,000 times consecutively, without any appreciable wear being indicated by a gauge applied to the levers and the key, both before and after this alternate action. Mr. Owen concluded by expressing his individual opinion that Chubb’s lock had never been picked. “The detector was the main feature of its excellence; and additional precaution, therefore, was only departing from its simplicity, and adding to the expense, without any commensurate advantage.”

In a subsequent chapter the degree of security afforded by various descriptions of locks, and the obstacles which they present of being picked, will come under notice; we therefore now proceed to describe briefly a few other tumbler-locks, or application of the tumbler-principle.

In Mr. Somerford’s lock, for which the Society of Arts gave a premium in 1818, an attempt was made to improve upon the ordinary action of tumblers. In most such locks, all the tumblers must ascend, although to different heights, before the stud of the bolt can pass through the slots; “which arrangement,” says Mr. Somerford, “gives an opportunity of introducing a nail, or a piece of stout wire, into the lock, and thus raising the tumblers without the necessity of using the key.” In his new lock, however, he made one lever to ascend while the other descended, by a somewhat complicated arrangement of slotted plates above and below the bolt. The key was so perforated as to be much endangered in respect to strength.

In Davis’s lock there is a double chamber with wards on the side of the key-hole. The key is inserted into the first chamber and turned a quarter round; it is then pushed forward into the inner chamber, where there is a rotating plate containing a series of small pins or studs, which are laid hold of by the key. By turning the key, the plate is moved round, the tumbler is raised, and the bolt is shot backwards and forwards. This lock, which is somewhat expensive, is used to some extent on Cabinet despatch-boxes.

The lock invented by Mr. Nettlefold is so constructed, that when the bolt is shot out by the key, two teeth or quadrants are projected from the sides of the bolt, which take a firm hold of the plate fixed on the door-post or edge. This construction is said to answer well for sliding-doors.

Mr. Alfred Ainger, in 1820, received a silver medal from the Society of Arts for a draw-back spring latch, in which the objects proposed were the two following--to render the lock more difficult of violation by a pick than those ordinarily in use; and to apply to it a key of which no ordinary person could take an impress, and which would be difficult of access even in a workman’s hand. The key is very peculiar; its pipe consists of three divisions, the section of the upper and lower divisions being circular, and that of the middle division triangular; the triangular portion is intended to give motion to some part of the interior of the lock during the rotation of the key. There are collars fixed on the extremity of the key, to act each on one tumbler; and there are modes, by varying the arrangement of these collars on an octagonal stem, to give something like a permutation to the number of variations to which the action of the key may be subject. The notches or slots are rather in the bolt than in the tumblers; and there are many peculiarities in the general arrangement.

In a lock invented and patented by Mr. Parsons, the tumblers are of a particular form, being hinged on a pivot at their centres, and working into and out of two notches cut in the under side of the bolt. It must be obvious that many variations in the adjustment of the tumblers of locks might be made, without vitiating the principle on which the action depends.

Many inventors have tried the use of an expanding web to the key, so planned that if the step of the web be long enough to reach the tumbler, it would be too long to pass through the key-hole; and therefore a principle of safety would operate by enabling the key to adjust itself at one moment to the size of the key-hole, and at another to the height of the tumbler. Mr. Machin of Wolverhampton invented such a key in 1827. The web of the key is movable on a countersunk pin, on which it can so far slide as to be drawn one-eighth of an inch from the barrel. The key-hole is of such a size as to admit the key only when the web is pressed close up to the barrel. When the key in this state is introduced, and is begun to be turned round, one of the notches in the web works into a raised circular edge of steel, placed eccentrically with regard to the lock-pin; so that as the key is turned, the web becomes drawn out, and is at its greatest elongation when it arrives at the tumblers: in the second half of its circular movement, the key becomes contracted to its original dimensions, and can then be removed from the lock.

Another mode of modifying the key has been introduced by Mr. Mackinnon, the object being to enable any person to change at will the pattern or arrangement of the movable parts of a lock and key; or to keep the key, when not actually in use, in such a state as to render it unavailing to any one but himself. It was a complex arrangement, which does not seem to have come much into use.

The lock invented by Mr. Williams, in 1839, may be designated a pin-lock, involving a principle analogous in many points to that of the Egyptian lock. This lock has a series of pins which reach through the cap, and are pressed to their places with a key like a comb or a rake-head. On the inner end of each pin is a flat piece of steel, in which is cut a notch for the passage of the bolt; but this passage is not clear until the notches in all the pieces of steel are in a right line. The pins are movable, and can be pushed either too far or not far enough to bring about the coincidence of position in the notches; and on this ground they are “double-acting.” Now the teeth of the key are of irregular lengths, each having a length just suited for pushing the pin to the proper depth: any other lengths of teeth would fail to open the lock. There is a mechanism of springs and levers to shoot the bolt when the pins in the plate are rightly adjusted. The arrangements in respect to the key are singular and somewhat awkward. The teeth which lock the bolt are not the same as those which unlock it, the user having to change ends and adjust the bit to a socket-handle. This is one among many examples in which a lock embodies several principles, the inventor having set himself the task of combining the excellences of many diverse locks.

In respect to the tumbler-locks generally, the simplicity of action, the strength of construction, and the non-liability of disarrangement, have given them a high place among safety-locks. The only danger seemed to be, that any person once obtaining possession of the key could take an impression from it, and thence form a key which would command the lock. Attempts have been occasionally made to obviate this danger, by supplying the key with movable bits which could be changed at pleasure, so as to constitute any number of effectively different bits in succession. But the locks being so constructed that the bolt could only be moved when the tumblers were in a certain position, the owner was placed in this predicament: that it was useless to alter the arrangement of the bits in the key, unless the tumblers were altered in a corresponding manner; and this would entail the removal of the lock from the door, and the re-arrangement of the interior mechanism.

One of the great defects of tumbler-locks made previously to the last ten years was, that the tumblers, when lying at rest in the lock, presented at their _bellies_ or lower edges precisely the same arrangement as the steps of the key. Indeed, in many locks of the present day, a good idea of the form of the key may be gained by feeling the bellies of the tumblers. The bellies are in fact cut out so as to compensate for the circular motion of the key, to allow them to remain at rest while the stump is passing through the gating. Even in tumbler-locks of the best construction the tumblers will vibrate more or less during the motion of the key; a defect which must be provided against in adjusting the lock, or the stump will be caught in its passage through the gating. Mr. Hobbs provides a simple remedy by enlarging the back part of the gating, the effect of which is as follows: when, in shooting back the bolt, as in unlocking, the key has got to its highest point, the stump enters the narrow end of the gating; but in shooting the bolt forward, as in locking, the stump enters the gating before the key has got to its highest point, and to allow for the slight vibratory motion of the tumblers during the passage of the stump, the gating is widened. The usual method of adjustment is to alter the forms of the bellies of the tumblers, thus greatly risking the security of the lock, a defect which was clearly perceived by Bramah [see pp. 67-70], and was one of the reasons which induced him to construct locks with slides instead of tumblers.

American locks on the tumbler-principle, and the relation which all such locks bear to the Bramah lock, will be better understood after the details of the following chapter.