Badge of the 14th Company Grenadier GuardsBadge of the 14th Company Grenadier Guards.
Badge of the 14th Company Grenadier Guards.
Badge of the 14th Company Grenadier Guards.
All the spring padlocks hitherto described have the defect of being in two parts; the bolt, being entirely detached from the tube when open, was liable to be lost, and to remedy this defect, modifications were introduced by which the bolt became a fixture in the tube and was opened by means of a key.
Fig. 35C, Plate V., is a lock which I found attached to one of my gates at Rushmore, in South Wilts. Externally, it exactly resembles the spring fetterlock, but within, the bolt which fixes the semicircular bar in its position when locked, is retained there by a spiral spring. To unlock it, a key with a female screw is put in at the end in the same position as the key of the Roman lock, and after seizing the male screw within, the bolt is screwed back against the spring, thereby releasing the semicircular bar or staple, which is then turned upon its hinge and drawn out of the opening on the side of the tube.
Fig. 36C, Plate V., is a precisely similar lock from Paris. Fig. 37C, Plate V., is another from Germany. Our modern handcuffs retain the form of the fetterlock, having the tubular case for the lock, which otherwise is not precisely the form most suitable to fit the human wrist. Fig. 38C, Plate V., is a section of an old handcuff obtained in Wiltshire, the bolt of which is forced out of the eye, not by means of a screw, but by a key of the ordinary form of a door key, inserted in the side of the tube, which when turned forces the bolt back against the spiral spring and releases the semicircular bar.
Whilst in some of the more modern contrivances the external form of the Roman spring padlock was retained, the interior mechanism having undergone changes, in others the interior mechanism is retained, the external form having adapted itself to the more modern uses. Figs. 39C, 40C, and 41C, Plate VI., is an old padlock which I obtained in Paris, the date of which I have been unable to determine, but a precisely similar one is attached to the iron chest of the Royal Society, which was presented to the Society in the year 1665, and for the knowledge of which I am indebted to Dr.John Evans, F.R.S., the Treasurer of the Society. Externally it resembles the modern padlock, but both ends of the semicircular staple are provided with springs on the principle of the Roman padlock. It is opened by means of a revolving key of modern form, which is inserted into the side of the padlock, and which, when given a quarter turn presses back the three springs upon the bolts, and the staple is then withdrawn bodily from the body of the lock. In this case, the staple, being quite separate from the lock, would be liable to be lost, as with the spring of the Roman padlock; so to remedy this defect we see in figs. 42Cand 43C, Plate VI., obtained at Clermont-Ferrand, in Auvergne, an improvement in which one of the arms of the staple passes down through the padlock and out at the bottom of it, where it terminates in a button, intended to prevent its being drawn entirely out and separated from the lock. The other arm is furnished with a spring as in the last example and, like it, is opened by a revolving key. When the spring is pressed back it is drawn out and merely turned upon its longer arm as a pivot.
Up to this point I have endeavoured to trace the gradual development of the European padlock from the earliest contrivance of Roman times up to the present time. In order to show its distribution and the varieties it has undergone in other parts of the world we must now return to the spring padlock in its earliest form. Figs. 44C, 45C, and 46C, Plate VI., represents an iron padlock from the Gate of Moultan, in India, now in the India Museum. It is in all respects similar to the Roman lock shown in figs. 21Cto 23C, Plate V., and needs no further description. Figs. 47Cand 48C, Plate VI., is a padlock obtained by me of a vendor of old iron in the streets of Cairo in 1881. It is constructed on precisely the same principle as the last, and is opened by a key thrust in longitudinally at the end of the tube, like the Roman key, but the opposite end of the bolt instead of being guided by a ring slipping along the parallel bar of the lock is curved round and inserted into a tube or socket in the parallel bar, like the Russian specimen and that from Swanscombe, in Kent. Figs. 49Cand 50C, Plate VI., is another specimen obtained by me at Cairo; it also resembles the Roman lock in its construction, except that the key instead of being thrust in at the end of the tube is put in underneath at right angles to the tube, and having enclosed the springs by means of an opening cut in the side of the key, in order to compress them, it is thrust sideways along the tube, the neck being guided by a slit along the bottom of the tube. Figs. 51Cand 52C, Plate VI., shows another specimen in my possession from India, which so precisely resembles the last that one is tempted to suppose they must both have been made in the same place, were it not for certain peculiarities which identify it as Indian. The key in closing on the springs is guided by two slits along the bottom of the tube instead of one, and beneath the tube is a projecting piece in the form of a Greek cross which fits into corresponding slits in the key so that none but the proper key can pass by it to compress the springs. This contrivance is therefore of the nature of a ward. Figs. 53C, 54C, and 55C, Plate VI., is another from India, now in the India Museum, the locality of which, viz., Myhere, is attached to it. Figs. 56Cand 57C, Plate VI., is an Egyptian manacle in my collection fastened in the same manner. Figs. 58C, 59C, and 60C, Plate VII., is a similar lock from Abyssinia, now in the British Museum, affording additional evidence that the key, with the lateral movement inserted at right angles to the lock, is African as well as Indian. Two padlocks precisely similar to this are in my collection from Mogadore, on the West Coast of Africa, having on them the peculiar Moorish ornamentation in brass which is characteristic of that country.
We have now to go to China for evidence of the continued distribution eastwards of this particular kind of spring padlock with the lateral key. Figs. 61C, 62C, and 63C, Plate VII., is a brass Chinese padlock and key in my collection. To the north of India we have figs. 64C, 65C, 66C, 67C, Plate VII., representing a padlock from Yarkand obtained by SirDouglas Forsyth, and now in the India Museum. It has also the key with the lateral action. Mr.Thomas Wrightsays that he possesses a similar padlock, given him by the British Vice-Consul at Jacmel, and obtained from Hayti, which he says was probably a century old, and either made in one of the Spanish colonies or imported from Spain.[31]SirGardner Wilkinsonalso mentions one from Meroe Island, in Egypt,[32]and Mr.H. Syer Cumingspeaks of one as having been obtained in Western Africa, but the locality is not stated.[33]
In order to show the modifications that this lock has undergone during its eastern migrations, I have represented (figs. 68C, 69C, and 70C, Plate VII.) a steel lock from Indore, India. It is furnished with a staple with two arms like the European specimen, fig. 39C, Plate VI., one of which only has springs attached to it; it is now in the India Museum. Figs. 71C, 72C, 73C, 74C, Plate VII.—also in the India Museum: the bolt with its springs is attached to plates forming an outside casing to one side of the lock, by which means the opening is concealed, and the opening for the key is also concealed in a casing for the other side, and opens also with a catch spring released by the pressure of a straight pin or wire introduced through a hole beneath the lock. Figs. 75C, 76C, and 77C, Plate VIII., is another variety, from Burmah; the key is introduced at the end of the tube by means of a male screw, formed somewhat like the propeller of a screw steamship. This screw is merely for the purpose of introducing the key into the tube by a half-twist; once in, it is pressed straight forward, and compresses the springs in the usual manner. Fig. 78C, Plate VIII., is the opening and key of a similar lock obtained by me in Nuremberg. It is constructed precisely on the same principle as the last, and with a similar object; it has all the appearance of being European, but I have no certain evidence that it may not have been imported from India. In figs. 79C, 80C, and 81C, Plate VIII., from Indore, India, we see the screw principle developed. Whether this originated in a lock of the last-mentioned form—and the screw, from having at first been used as a ward, was ultimately employed to release the bolt by a screw motion—I know not; but it exactly resembles in its construction the lock shown in fig. 35C, Plate V., from the gate at Rushmore, Wilts, and those of like form from France and Germany already figured and described. The bolt is retained in its place when locked by a spiral spring, and withdrawn by a screw key inserted at the end. Whether this is an independent growth in the two hemispheres, or copied the one from the other, I have no present means of determining. Unfortunately, when the objects in the India Museum at South Kensington were transferred from the old India Museum their history was lost; but I have figured none except those which have the localities attached to them. Figs. 82C, 83C, and 84C, Plate VIII., is a steel lock from India of similar external form to fig. 71C, Plate VII.; but the screw principle appears here to have entirely superseded the spring, which is altogether wanting, and it is dependent for its action entirely on a screw key inserted at the end, and by means of which the bolt (which itself formed the staple) is screwed up or screwed back again as required. As a parallel to this, the specimen in my collection represented in figs. 85C, 86C, 87C, 88C, and 89C, Plate VIII., may be given. It was obtained by me in Brussels, and resembles the tubular lock only in external form. The staple is secured to the tube at each end by eyes let into the side of the tube, through which a pin is passed, and screwed up or unscrewed by a key put in at the end of the tube. When unscrewed the pin is withdrawn and the staple taken out bodily. In this, as in the Indian specimen last described, the original spring mechanism has entirely disappeared; but, although resembling each other in this respect, there is nothing analogous in the two systems, which, from differences in the details of their construction, appear to be quite independent contrivances. Figs. 90C, 91C, and 92C, Plate VIII., represent a padlock and key from Toomkoor, Mysore, India. It is a barbed spring padlock of the ordinary kind, but the springs are closed preparatory to being withdrawn by means of a common revolving key inserted in the side and having a broad slit in the middle of the revolving plate. By giving the key a quarter-turn the slit in the key-plate compresses the springs, and they are then withdrawn from the lock. The action of the key in this specimen resembles exactly that of the padlock from Paris (fig. 39C, Plate VI.) and that of the Royal Society chest, except that in the Paris and Royal Society specimens two springs are compressed by means of a solid plate, whilst in the Toomkoor example a single-barbed spring is compressed by the action of a slit in the key. Barbed tubular spring locks of precisely the same form as the Chinese ones are also used in Japan, of which figs. 93C, 94C, 95C, and 96C, Plate IX., represent a specimen in my collection. Of these, some of the keys entered at the end of the tube; others are put in at the side, as shown in fig. 97C, Plate IX. The key, which, like the lock, is of brass, is placed in a handle, which shuts up like the handle of a knife (as shown in fig. 96C, Plate IX.) for convenience of transport. Another specimen from Japan (represented in figs. 98Cand 99C, Plate IX.) resembles exactly the Toomkoor specimen from India, the springs being compressed by means of a revolving key. This must certainly be regarded as the first stage of improvement upon the original Roman lock, and its employment in Europe, India, and Japan is noteworthy.
Amongst the specimens of these tubular spring locks, which appear to show evidence of connection over wide areas, are those which are constructed in the forms of animals. Figs. 100Cand 101C, Plate IX., is a representation of a bronze padlock in the form of a fish, now in the Louvre, at Paris, figured byM. Liger. It is there described, though not without hesitation, as an Egyptian lock; if so, it is probably of the Romano-Egyptian period: the springs enter at the mouth of the fish, and are released by a key put in at the tail. Figs. 102C, 103C, 104C, Plate IX., represents a precisely similar fish-shaped padlock of iron from India, and now in the India Museum. Figs. 105Cand 106C, Plate IX., is a Roman bronze lock in the form of a lion or horse, in the possession of Dr.John Evans, F.R.S., and here copied by his permission; a similar one is in the British Museum. Figs. 107C, 108C, and 109C, Plate IX., is another, also in the form of a lion, and about the same size, from China, in the collection of Mr.Chubb, the well-known locksmith. In all these the springs enter at the stern of the animal, and the other end of the bolt turns up and back in the form of a tail, and enters the neck of the animal behind the head. The key in the Chinese specimen has a peculiar secret contrivance to prevent its being inserted in the hole for it by anyone not acquainted with its construction. The head of the key will not enter the keyhole unless the handle end is put in first and slipped along the shank of the key, as represented in the drawing, fig. 109C, Plate IX. Mr.Romilly Allen, whose work on Scotch wooden tumbler locks I have already quoted, refers incidentally in his paper to spring locks, and says that he has himself seen them used in Persia in the forms of animals. We are thus led to infer that the practice of making them in these forms may have existed, or may still exist, continuously throughout the region referred to, and that, like the mechanism itself, and like many other articles of commerce, they may have passed by traffic from place to place, and been copied and adopted in the localities in which they are found. Fig. 110C, Plate IX., is a padlock obtained by me at Cairo; similar ones are in common use on out-houses at Naples, the long bar at the top denoting its descent from the Roman padlock, although the construction of the lock is different.
We now come to the principle of the revolving key in common use at the present time. It has been already shown that in using the Roman lock (figs. 15Bto 20B, Plate III.) the part of the key containing the pins had to be put in vertically, and then turned a quarter circle, so as to bring the teeth horizontally beneath the tumblers previously to lifting them. It is possible that this may have suggested the first idea of employing the twist thus given to the key to the shooting of the bolt. Fig. III, Plate IX., taken fromM. Liger'swork,[34]represents a Roman key found in London; it has a plate furnished with teeth, evidently intended to raise tumblers, and the stem of the key is piped for the purpose of fitting into a broach or pin, so that the plate with the teeth, when the key is turned round on its pivot, may fit into its proper place beneath the bolt and raise up the tumblers. Fig. 112, Plate IX., is a drawing of another key similarly formed, having two teeth and a piped stem; it was found in Lothbury, in London, 16 feet beneath the surface, and is figured in Mr.Syer Cuming'spaper on keys in the 'Journal of the Archæological Association.'[35]These keys appear hardly to admit of any doubt as to their mode of use, and may therefore be regarded as the earliest specimen of revolving keys, although applied to a different purpose from the revolving key of our own time. The most primitive kind of lock with a revolving key that I have met with is one represented in figs. 113, 114, 115, 116, Plate X. It is from India, and is in the India Museum. The key is applied to a square vertical tumbler of the Scandinavian type with two arms to fit into two notches in the bolt; the lower end of the tumbler terminates behind the bolt, in a semicircular form; the key, when turned upon its broach or pin, as the case may be, impinges upon the sides of the semicircular portion and raises the tumbler out of the notches on the top of the bolt, and afterwards the end of the key-plate passes into one of a series of notches on the under side of the bolt and moves it, whilst the tumbler is, at the same time, raised clear of the bolt. The key being turned several times continues the movement, pushing the key forward notch after notch, until the tumbler again falls into other holes provided for it, and keeps the bolt secure. All here is of wood, except the key, which is of metal, and it is provided with slits to pass the wards, adjusted to them in the revolution of the key-plate upon its pivot. It might be supposed from this that it was a modern adaptation to an ancient system of vertical tumblers, had not a very similar, but simpler, lock existed in China. The drawing (figs. 117, 118, 119, 120, 121, Plate X.) of a Chinese lock was kindly sent me by Mr.Romilly Allen. In this specimen the bolt is shot in nearly the same manner as the last specimen, but the tumblers are raised independently by means of a T-shaped key (fig. 121, Plate X.), similar to that used with the Scandinavian lock (fig. 3C, Plate IV.). The key from the outside is put into the vertical slit between the tumblers, when it is turned a quarter circle so as to bring the arms of the T in a horizontal plane. It is then pressed back, when the returns of the T enter notches provided for them in the tumblers. The tumblers are then raised, and the key or handle,a, turned. From the inside the tumblers are raised with the two fingers before shooting the bolt.
M. Ligersupposes that the lifting key of the Roman lock was of Asiatic origin, and that the revolving key came into use amongst the Romans about the commencement[36]of our era, and many of the keys from Pompeii are constructed on this principle having slits for the passage of wards. Fig. 122, Plate X., is a Roman key of this kind in my collection. The ward system came into general use afterwards and was much relied upon to the exclusion of others in the Middle Ages. The ward system may be defined as a system of lock in which obstructions are placed to prevent any but the proper key from entering to turn the bolt; as such it is distinct from the tumbler system, in which security depends on obstruction introduced to prevent the bolt from being drawn by the key. The tumbler is, in fact, a bolt of a bolt. Reference to fig. 10B, Plate II., representing the Egyptian lock, will show that besides the two pins with which the key is provided for lifting the tumblers, there is a pin attached to the under side of the lock opening, which enters a hole in the key. This is of the nature of a ward, since none but a key with a hole in the proper place could be raised up high enough to lift the tumblers clear of the holes in the bolt. Mr.Romilly Allenalso mentions that in one of the Scotch locks from Snizort, a notch is placed in the key and a corresponding pin in the lock, to prevent the lock from being picked, and that the key-hole is divided by a thin iron plate which is the only thing approaching a ward that appears in any of the wooden locks of Scotland. The peculiar shape of the tumblers and tumbler-holes in the bolts of the Roman lock, already described, with teeth made especially to fit them, must be regarded as a kind of ward, although applied to tumblers, since their object is to prevent any but the proper form of key from entering.
The further development of the ward-system in the Roman tumbler-locks, though it certainly existed, is involved in uncertainty, since none of the wards appear to have been preserved, but the fact of some kind of ward having been used is evident from the slits in the keys represented in fig. 122, Plate X., which are of common occurrence. The cross-shaped wards beneath the Indian spring padlock already described in connection with figs. 53C, 54C, and 55C, Plate VI., must certainly be considered to be wards, although open to view, and not concealed beneath the lock-plate. There are also found in association with Roman remains, keys of which fig. 123, Plate X., from Chalons, fig. 124, Plate X., from the Museum at Saumur, and fig. 125, Plate X., from the Museum at Saint Germain, are examples.[37]These keys so greatly resemble the Asiatic keys used with the spring padlock, that it is difficult to believe they were not employed in the same way, but as they also resemble the Roman perforated plates of the tumbler-lock keys that are provided with teeth, it is probable they may have been intended for raising tumblers in some way not yet explained. No tubular spring lock adapted to be opened with a key inserted underneath, and opened with a lateral movement like the Indian and Egyptian ones, has to my knowledge been found amongst Roman remains. Fig. 126, Plate X., is a modern English latch-key of similar form, furnished with a ward-plate and used for raising a common latch: they are now generally disused, from being unsafe. With the revolving keys resembling the modern form, found at Pompeii and elsewhere, slits for fixed wards are common, and show that the Roman keys of the commencement of the present era resembled our own. During the Middle Ages reliance was placed almost entirely on the ward system, and many complicated contrivances were introduced, of which fig. 127, Plate X., is a specimen, until the close of the last century, when their insecurity led to the re-introduction of tumbler-locks.
It is not known exactly when this took place, but probably at some time during the 18th century, and possibly earlier. This time, the tumblers instead of being vertical (as was the case during what may be called the early tumbler period) were horizontal, resting on a pivot above the bolt and kept down by a spring. Figs. 128, 129, and 130, Plate X., is a tumbler lock in the possession of Mr.Chubb, found whilst repairing an old house at Funtley, Hants, said to be 200 years old. If so it must be regarded as the earliest specimen of the second tumbler period. The tumbler moves on a pivot, and is kept down by a spring, the revolving key raises the tumbler by pressing up the curved bar attached to it, which raises the stud of the tumbler out of the notch provided for it on the upper side of the bolt, thereby freeing the bolt, so that by further turning the key it is enabled to shoot the bolt. The tumbler, it will be seen, cannot be raised too high. If the plate of the key is long enough to raise the stud of the tumbler out of the notch, a key with a longer plate will answer the same purpose. To remedy this defect and necessitate the employment of a key of exactly the proper size, Mr.Barron, about the year 1778, introduced an improvement known by his name, represented in fig. 131, Plate X., in which the bolt is provided with a slit along the middle just wide enough to allow the stud to pass; the slit has notches both above and below, so that if the stud is raised too high by a key with too long a plate it is forced into the upper notch and the bolt continues immovable. He also introduced two tumblers requiring to be raised to different heights in order to coincide with the slit in the bolt by means of different projections on the edge of the key plate, so that the bolt could only be shot by means of a key with a plate expressly constructed to fit the lock, and having two projections of the requisite length. This principle of employing two or more tumblers is the one on which nearly all subsequent improvements have been effected. Those who desire to prosecute the subject further will find a variety of modern tumbler locks in my collection introduced during the latter half of the last and commencement of the present century. They are all, in the main, varieties of one principle, terminating in theChubbandHobbslocks of the present time. As this paper relates only to primitive locks I do propose to describe them here. The continuity which pervades all the ramifications of the modern lock is not less complete than in the earlier forms, and would well bear treating in the same manner as those which I have described. The Bramah lock, though in external appearance differing from the others, is no less based upon the earlier forms, and may be described as a union between thewardand thetumblersystems. It is a ward system, because the obstructions introduced into the mechanism are intended to prevent the turning of the key to shoot the bolt by any but a key of the proper construction. It is a tumbler system because the impediments so placed upon the turning of the key are in fact tumblers packed round the cylinder of the key (retained by springs), and allowing the passage of the key-plate only when pressed down to the various depths to which each separate tumbler is adapted in order to provide an open passage for the key-plate all round. This union of ideas developed separately in different branches of the same trade, device or industry, corresponds to the crossing of individuals and breeds in nature, which is so necessary to reproduction. The analogy, as I have already intimated elsewhere, might be carried even further and closer if space permitted. It is a necessary condition of the absence of creative power in nature, and applies equally to all the processes of evolution whether of species or of ideas, but the subject requires broader treatment than can be given to it here. My object in writing this paper being to trace the development of particular forms rather than to generalise, I must leave the philosophy of the subject for separate treatment.
From the foregoing description of the various kinds of primitive locks in use in different countries it will, I think, have been made evident that some of them most certainly have been derived from a common centre. The wooden key-drawn pin-locks have spread over the region extending from Egypt to Yarkand. The Scandinavian wooden locks of the same kind, though differing in the details of their construction, we have seen are common to Norway and Scotland, and by some means have been carried to the West Indies and British Guiana, whilst the tubular spring padlock of the Roman age in Europe is the same that is found throughout the whole region extending from Italy to China and Japan on the east, northward into England and Scandinavia, southward into Abyssinia, and westward into West Africa and Algeria, Spain, and on as far as the West Indies.
It is sometimes thought when simple contrivances such as weapons of stone and bronze, some of the simpler kinds of ornaments, and of tools obviously adapted to primeval life are found to extend over wide areas, and in places very remote from one another, that the few ideas necessary for the construction and use of them might easily have suggested themselves independently in different places. To the student of primitive culture who has become impressed with the persistency of art forms, this independent origin of such things does not appear so certain even in the case of the most simple contrivances. But when we come to a complex piece of mechanism, such as a spring padlock having several parts—the spring, the case, the parallel bar, and the key, in all of which the resemblance is maintained in distant countries, and which, with slight modification and continuously progressive improvements, are put together in the same manner in all parts of the world—such a supposition cannot be admitted, the necessity for a common origin is apparent, and the study of the periods and the circumstances connected with the distribution of it cannot be set aside as superfluous.
Assuming that the tumbler pin-lock and the spring padlock cannot be traced back earlier in Europe than the commencement of our era, it is by no means certain that they may not have existed earlier elsewhere. The commerce carried on with the East in early times was of a nature to render it very probable that any contrivance for securing goods should have spread from place to place with the merchandise exported and imported between China, India, and Europe. A brief survey of the trade relations between different countries will be sufficient to show this.
The expedition ofAlexandergave rise to intercourse which was kept up by the Greek kingdom of Bactria, and recent Indian discoveries both of coins and sculptures prove more and more the great influence which Greek art exercised in India up to the commencement of our era.Strabosays that, aboutB.C.22,Nicolaus Damascenusfell in with three Indian ambassadors at Antioch Epidaphne on their way to the Court ofAugustus, and that their credentials were in the Greek language.DiodorusquotingIambulusspeaks of KingPalibothrain the early part of the 1st century as a lover of the Greeks.Dio Chrysostommentions that the poems ofHomerwere sung by the Indians, andÆliansays that not only the Indians but the kings of Persia translated and sang them. If the travels ofApolloniusandDamisare to be credited, the Greek language was spoken in the Punjaub in the first half-century of our era, and frequent intercourse appears to have taken place between that country and Egypt.[38]Plinyin the 1st centuryA.D.says, on the authority ofVarro, that under the direction ofPompeyit was ascertained that it took seven days to go from India to the River Icarus, believed to be the modern Roscha, in the country of the Bactri, which discharges itself into the Oxus, and that the merchandise of India being conveyed from it through the Caspian Sea into the Cyrus, might be brought by land to Phasis in Pontus in five days at most.[39]The best steel used in Rome was imported from China.[40]Arrian, in the 2nd centuryA.D., speaks of a frequented way, λεωϕóρος, extending in the direction of India through Bactria; after which four embassies from the East are noticed by ancient writers, one toTrajan,A.D.107; another toAntoninus Pius,A.D.138-161; a third toJulian,A.D.361; and the fourth toJustinian,A.D.530. These are but scant memorials of an intercourse which must have been frequent between India and Rome, and which reached its highest development during the reigns ofSeverusandCaracalla, in the commencement of the 3rd centuryA.D.
Turning now to the southern route of communication with India,Plinydescribes Taprobane (Ceylon), and mentions an embassy sent from thence to the EmperorClaudius. The discovery of the monsoons during the 1st century was the means of creating a great trade between India and Alexandria.Strabosays that in the time of thePtolemiessome 20 ships only ventured upon the Indian seas, but that this traffic had so greatly increased that he himself saw at Myos Hormos, on the Arabian Gulf, 120 ships destined for India.Plinygives in detail the route from Alexandria to India in his time, and says that it was well worthy of notice because in each year India drained the empire of at least 550 sestertii, estimated at £1,400,000 of English money, giving back in exchange her own wares, which were sold at fully one hundred times their original cost, and he says that the voyage was made every year by the following route:—Two miles distant from Alexandria was the town of Juliopolis, supposed to be Nicopolis. The distance from thence to Coptos up the Nile was 308 miles, and the voyage was performed with a favourable wind in 12 days. From Coptos the journey was made on camels to Berenice, a seaport on the southern frontier of Egypt, 257 miles, in another 12 days. Here the passengers generally set sail at midsummer, and in about 30 days arrived at Ocelis, in Arabia, now called Gehla, or at Cane, supposed to be Cava Canim Bay. From hence, if the wind called hippaulus happened to be blowing, it was possible to arrive at Muzitis, the modern Mangalore, which was the nearest point in India, in 40 days. This, however, was not a convenient port for disembarking, and Barace was therefore preferred. To this place pepper was carried down in dug-out canoes made out of a single trunk from Cottonara, supposed to be Cochin or Travancore. The return voyage was usually made in January, taking advantage of the south-east monsoon, by which means they were able to go and return the same year. But whenPlinywrote, the trade with India was only in its infancy, afterwards Greek factories were probably established at the Indian seaboards, which accounts for the Greek names for some of the towns on that coast.
But the people of Alexandria having become insolent in their prosperity,Hadrianwas led to encourage the route through Palmyra, which was the most direct road to India. Even in the 2nd centuryA.D.the trade between Rome and India through Palmyra must have been considerable, for it drew the attention of the Chinese. Their annals speak of it as carried on principally by sea; they mention Roman merchants in relations of commerce with and visiting Burmah, Tonquin, and Cochin China, and they have preserved the memory of an embassy from the Roman emperor, which in the yearA.D.166 was received by the Chinese sovereign. Arab or native vessels appear to have brought the produce of India up the Persian Gulf to the mouth of the Euphrates. At Teredon they discharged their cargoes, and the merchandise was then carried to Vologesia by camels; at this place the merchants of Palmyra took it up and it was here exchanged for the produce of Europe. Even as late as the 5th century, ships from India and China are mentioned lying at Hira on the Euphrates, a little to the south of Babylon. Through the influence of this trade Palmyra grew rapidly into wealth and power until the widow ofGalberiusthrew off her allegiance to Rome. This led to the destruction of the city byAurelian,A.D.275, which put an end to the Roman trade with India through the Persian Gulf. The Alexandrian trade with India fell off about the same time, and the barbarians occupied Coptos, the port of embarkation for India, aboutA.D.279.
After the fall of Palmyra the Indian trade was transferred to Batne, near the Euphrates, but it lasted only a short time, and in the 4th and 5th centuries may be regarded as having become extinct in so far as Roman merchants were concerned. The trade, however, was still kept up by the Arabs.Epiphanius, aboutA.D.375, gives an account of trade carried on through Berenice, by which the merchants of India imported their goods into the Roman territory, and there is also Chinese authority for believing that a great trade between Rome and India existed in the 6th century.Ma-Touan-Lin,A.D.1317, in his researches into antiquity, affirms that inA.D.500-516 India carried on a considerable commerce by sea with Ta-Tsin, the Roman Empire, and with the Ansi the Syrians,[41]but Arab and not Roman vessels were employed.Masoudisays that in the early part of the 7th century the Indian and Chinese trade with Babylon was principally in the hands of the Indians and Chinese. The usual passage after rounding the Point de Galle was to creep up the Madras coast during the S.W. monsoon and take a point of departure from Masulapatam towards the leading opening of the Ganges.[42]Meanwhile the overland trade between Europe and India in the 3rd and 4th centuries was carried on by the Sassanidæ, who in the 4th century entered into commercial relations with China, to which country they sent frequent embassies in the 6th century, and through this route silk was imported into Europe. InA.D.712 Sind was conquered by the Arabs, and in addition to the kingdom of Mansurah and Multan, other independent Muslim governments were established at Bania and Kasdar.[43]There is also the evidence of the merchantSulaminand the researches of Mr.Edward Thomasinto the coins of the Balhara to prove the continuance of Arab intercourse with India during the 9th century.
During all this time the relations between Scandinavia and Rome appear to have been scarcely less extensive. Although the Romans never succeeded in penetrating Scandinavia, the discovery of coins, vases in bronze and glass, and other objects of art, is sufficient to prove that Scandinavian art was greatly influenced by intercourse with Rome during the first part of the 2nd century of our era. In the early stages of society, communication by sea offered greater facilities for traffic than land journeys, and for this reason the Island of Gotland, now so isolated and rarely visited except by antiquaries, appears to have served as a portal for the entry of Roman and Oriental goods and civilization into Scandinavia.[44]After the fall of the Roman empire, Scandinavia was left to its own resources, aided by occasional intercourse with Byzantium, until in the later iron age, extending from the 8th century to the middle of the 11th century, another line of communication was established with the East, still entering Scandinavia mainly through the Island of Gotland. Mr.Hildebrandrecords the discovery of 20,000 Arab coins in Sweden and Gotland, and traces the channel of their transmission by Russian finds from the states near the Caspian, through Russia to the shores of the Baltic, and thence, thanks to the commerce established by the inhabitants of Gotland, over to that island. From Gotland, and probably also by direct intercourse with Russia, the Mahomedan coins were spread over Scandinavia, being more common in the eastern provinces of Sweden than in the west or in Norway. The greater part of these coins appear to have come into Sweden between the years 880 and 955, but the latest belongs to the year 1010. On the line of communication here indicated, iron keys of the kind adapted both to the tumbler lock and the spring padlock have been discovered in the governments of Vladimir and Jaroslav, in the graves of the Neriens,[45]dating about the 8th centuryA.D., showing that in all probability it was by this line that the use of these locks were imported into Sweden. The key of the padlock found here was of the form of the Roman key, (fig. 21C, Plate V.), the Indian one (fig. 46C, Plate VI.), and the modern one from Cairo (fig. 47C, Plate VI.). It also resembles that of the Swedish lock (fig. 26C, Plate V.), and belongs to the most primitive form of the mechanism.
Whilst this traffic was being carried on between Scandinavia and the East, the intercourse of the Vikings was kept up with Britain, Ireland, and the coasts of the English Channel, commencing in 787 and continuing to the 11th century. These Western relations, like those with the East, appear to have taken place chiefly through Gotland; and the number of Anglo-Saxon coins found in that Island and the East of Sweden greatly exceed those discovered in Norway and the West.
The foregoing summary of the evidence of commercial relations between Southern Europe and the East and North during the early part of the Christian Era is sufficient to show that ample facilities existed for the spread of early forms of locks and keys. The padlock, more especially—which, as I have said when referring to the etymology of the word "pad," was the class of lock associated with portable merchandise—must have been carried into all those parts of the world between which commercial relations had been established.
At what time and through what particular channels the various kinds of locks were distributed can only be determined after more extended inquiry into the archæology of padlocks. Some points may, however, I think be considered to be more or less established by the evidence I have adduced. The particular form of padlock represented in fig. 44C, Plate VI., from India, and fig. 21C, Plate V., from the Roman period of Europe, must in all probability have been communicated in Roman times, as I am not aware that this precise form of padlock was in use in Europe later than the Roman age, having been superseded by the more modern improvements which have been described in this paper. The use of padlocks in the forms of animals in Egypt, Persia, and China, must also very probably belong to the same period. The Chinese and Japanese padlocks appear to belong to a more advanced stage of the development of the mechanism, and correspond to the form used in Europe in the Middle Ages; whilst the use of the revolving key in Europe, India, and Japan, to compress the springs, as shown in figs. 39C, Plate VI., 90C, Plate VIII., and 98C, Plate XI., must date from a still later phase in the art; and unless they are to be regarded as improvements introduced independently in those countries, the idea must have spread by means of Arab traders, if not still more recently. In like manner, the adoption of the screw principle with these locks must either have been conveyed by traders, or applied independently in different countries to the form of padlock already in use. The hinge of the staple, as seen in figs. 26Cand 31C, Plate V., though derived from the earlier form of the parallel bar, which has a wide distribution, has not been universally adopted, but is used chiefly in Sweden and Europe, and is an improvement introduced, no doubt, in modern times. Further information is needed to enable us to trace the distribution of all these different varieties more continuously, before any satisfactory judgment can be formed as to the date of connection. In Scandinavia we find the padlock in use in Gotland, in Björkö, and in Sweden; andHans Hildebrand, in his work on 'The Industrial Arts of Scandinavia,'[46]published by the South Kensington Museum, says that they were already known in that region in Pagan times. It is to be hoped that this announcement may be only a prelude to some more detailed publication of his researches into a subject to which the present paper can only be regarded as a first introduction—not previously attempted, that I am aware of, in its ethnological and commercial bearings. Local archæologists must work out the rest. Enough has, I trust, been said to show that a large field lies open to the student of the archæology of locks and keys, and that whenever the history of this mechanism is traced in Scandinavia, Persia, India, and China, in the same way that I have endeavoured to trace it in Europe, much light will thereby be thrown on the ramifications of trade and the commercial relations of distant countries in non-historic times.
Fig. 1. Japanese book fastening derived from the common pin.
Figs. 2 and 3. Common wooden bolt used at Gastein, in Austria, at the present time.
Fig. 2. Front view.
Fig. 3. Transverse section on A B.
a.Handle.b.Bolt.c c.Slit for handle,a.
Figs. 4 and 5. Wooden bolt with pin fastening (supposed form).
Fig. 4. Front view.
Fig. 5. Transverse section on A B.
Figs. 6 to 8. Wooden single tumbler bolt (supposed form).
Fig. 6. Front view (open).
Fig. 7. Front view (closed).
Fig. 8. Transverse section on A B.
Figs. 9Ato 11A. Wooden double tumbler lock from the Faroe Islands.
Fig. 9A. Front view.
Fig. 10A. Longitudinal section.
Fig. 11A. Transverse section.
a.Bolt.b b.Teeth of key,c.d d.Tumblers.e e e.Block.f f.Holes in bolt.
Figs. 12Ato 17A. Old Scottish wooden tumbler lock (Patent Museum).
Fig. 12A. Front view.
Fig. 13A. Side view.
Fig. 14A. Longitudinal section.
Fig. 15A. Transverse section.
Fig. 16A. Section through A B.
Fig. 17A. Section through C D.