The Road Services in Kent in the Nineteenth Century

213 The road services of Kent in the nineteenth century f.w.g. andrews When the steam locomotive railway become a practical proposition, numerous voices were raised to lament the impending disappearance of the horse, but the reports of the horse’s death were much exaggerated, and this is very evident in a study of the changing patterns of the road transport services of Kent in the nineteenth century. Leaving aside the not very successful (if very significant) Canterbury and Whitstable railway, which opened in 1830, railways did not come to Kent until the 1840s: the South Eastern Railway’s main line from London to Dover opened in 1844, the very important branch through to the Thanet resorts following in 1846. By 1899 the railway pattern was all but complete, and by that date of the railways which criss-crossed Kent at the time of the 1923 Railway Act, only the East Kent Light Railway had not been built (not that it was very important when it did begin operation in the years just prior to 1914) and what was to become the Kent and East Sussex railway was under construction. In those six decades, therefore, the whole pattern of land transport in Kent was changed, and this paper is an attempt to consider the changes in the pattern of the road transport industry during those years, basing the analysis in the main on directory evidence. O f the various directories which exist for Kent in the nineteenth century two are used for this purpose, firstly that published by Pigot and Co. in 1839,1 and secondly that published by Kelly in 1899,2 and from an analysis of the entries in these two directories it is possible to appreciate in detail the changes that had come upon the scene of rural transport in those sixty years. There is certainly the problem of how far any directory can be regarded as complete and/or reliable. The Kelly directories were produced from material supplied by agents, and it is reasonable to assume that the Pigot directory must have been compiled in the same way. Though these local agents must have been given some sort of briefing as to what was required, the way the information was collected and later collated shows that there must have been a great deal of inconsistency in working practices. It F .W.G. ANDRE WS 214 seems reasonable to assume that an entry had to be paid for (though the present-day company which produces Kelly’s directories is not clear about this, and non-display entries in the Yellow Pages are free), and so not every operator will have been represented, and this must especially apply to the one-man carrier, the local Barkis. All the same, the directories are a mine of information, and probably contain more than they omit, so the exercise is worth while. R oad Transport in the late 1830s Coaches The Pigot directory sub-divides land transport into three groups – coaches, vans and carriers. No overall list is given of companies or operators which comprise any of these groups, so their overall numbers must be determined by a town-by-town collation of the various transport entries. The directory lists 48 identifiable coach services by the coach name, such as ‘The Balloon’, ‘The Tantivy’ and so on, and almost all of these relate to services connecting London with Kent. Some services, such as ‘The Defiance’ appear a number of times in the various town entries, as the passage of the coach through Kent is timetabled, others appear just once, though these tend to be journeys between major towns, rather than London services. Almost all the long-distance coaches ran a return service once daily, some, but by no means all, on a Sunday as well; the rest ran three services up and three balancing services down in the week, though the Paris and Dover Mail was recorded as running only twice a week in each direction. Nearer London, as for example at Lewisham (then part of Kent), Blackheath or even Gravesend, as the various services began to bunch together the coaches are described as running ‘continually’. Clearly some towns were very much better served than others (see Table 1). This table needs to be treated with some caution. Foots Cray’s figure refers to coaches passing through between London and Maidstone; anybody wishing to go to Foots Cray would board a London or a Maidstone coach, and the same is true of Dartford’s coaches. Maidstone’s coaches were certainly dominated by the London services – of the thirteen coaches which were advertised as starting from Maidstone, seven coaches left for London, but the coaches which had Maidstone as their destination had very varied start points, though in the main along the line of what is now the A2 trunk road. Canterbury’s outward bound traffic was dominated by services to London [9], to Dover [7], Sandwich and Deal [4], with a single trip each to Maidstone and Margate and Ramsgate. With the exception of services from Herne and Herne Bay [2], Maidstone [2], and Deal and Walmer [1], all services to Canterbury were from towns along the main THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 215 London-Dover road, the present A2, but though some towns had more services than others, the pattern of routes was pretty much the same for all – services along the main London-Dover routes via Maidstone or via the Medway towns were all-dominant. Outward services from Tonbridge and Tunbridge Wells were dominated by London traffic, via Sevenoaks [17] or that along the coast to Hastings and Brighton [5]. Services to Tunbridge Wells (no dedicated services to Tonbridge are shown) were all from towns along the route to London. The directory lists 90 services between pairs of towns which may be part of these London services, or separate routes along the line of the London routes. There appears, for example, to have been a service between Maidstone and the Medway towns seven times daily in each direction which may simply have been a record of the main services which passed between those towns, or it may be a dedicated Medway towns service. Others may be part of cross country routes operated by un-named coaches, such as that which ran daily between Maidstone and Tenterden, or the daily Deal and Walmer to Herne Bay service, which was operated in conjunction with the steam packet services to Herne Bay. Generally speaking though the Pigot directory suggests very strongly that unless you lived on a main road between London and one of the towns in Kent, your chance of having a convenient coach service was very poor. A possibly more reliable list of Kent coach services is that provided by TablBLe 1. NumberMBER OF COACH SERviCES FROM, OR TO, MAjOR TOWNS in 1839 [OnlNLy TOWNS wiTH six OR MORE TO OR FROM ENTRiES ARE listed here] Towns S ervices Towns (Ctd) S ervices To From To From Brighton 7 H astings 13 Bromley 11 Lamberhurst 8 C anterbury 12 22 LONDON 136 C hatham 6 17 Maidstone 18 13 D artford 2 21 Margate 6 6 D eal/ Walmer 7 6 R ochester 4 19 D eptford 10 S andwich 2 9 D over 44 12 Tenterden 8 2 F olkestone 7 3 Tonbridge/ Tunbridge Wells 8 29 F oots Cray 17 Wateringbury 6 Gravesend 3 13 F.W.G. ANDRE WS 216 A lan Bates,3 whose work is based on a study of the licence plates issued to coach operators in 1836, near enough in time to Pigot’s directory to be closely comparable. Bates lists 40 named coach services running from London into Kent, plus another seven un-named services and the two Royal Mail services, one to Dover and one to Hastings, via Sevenoaks, a total of 49, very similar to Pigot’s total. Bates’ information enabled him to indicate the number of passengers who could have travelled in each coach, a grand total of 4,406 a week, plus the Royal Mail coaches, for which no plates were issued, and so no figures of capacity available. He also listed the secondary coach operators, which included those who operated along part of the routes between London and other towns as well as truly cross-country routes. Since the individual vehicles are listed as well as their operators, this information makes it possible to calculate the number of coach seats available in any one week, a total of 6,260. Journeys listed are in one direction only, but as there must have been return services in all cases, the total figure [4,406 + 6,260 = 10,666] should be doubled to give 21,332 seats available. I f that result multiplied by 52 for weeks in the year an approximation of the total available coach seats in Kent in the year may be arrived at; it is 1,109,264. Not all seats will have been filled of course, but all the same it indicates that about a million people could have been travelling about in Kent each year on coaches alone. It is at first perhaps surprising that over 650,000 of those seats were available on the secondary services, but on closer examination the term secondary is seen to be rather misleading (Table 2). TableABLE 2. SEATS PER STARTiNG POiNT in THE SECONDARy COACH services, 1836 Start Point S eats % Start Point S eats % A shford 36 0.6 Gravesend 1,715 27.4 Brighton 144 2.3 H astings 24 0.4 C anterbury 1,583 25.3 Maidstone 947 15.1 D over 768 12.3 Margate 913 14.6 E. Peckham 36 0.6 R amsgate 42 0.7 E denbridge 28 0.4 S ittingbourne 24 0.4 A lmost all the routes from these towns go to other towns on the main coach routes from London, except that from Canterbury there were services to the Thanet resorts, and from Margate most of the services were very local ones to Ramsgate: there are few really cross-country services at all, other than a few oddities like Edenbridge to Westerham, or Canterbury to Lydd. Generally speaking, Bates’ evidence supports the conclusion already THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 217 drawn from the Pigot evidence, that is, unless you lived on a main road which was part of a route to London, there would not be a coach service to hand. In this respect the coach services were exactly replicated by the pattern of the first stages of the railway system – the lines were built to connect major centres, which in the south of England, meant connection with London. Vans The Pigot directory lists as a separate group of road transport operators those who provided van services. There appear to have been at least 107 distinct services run by the van operators who were at least 42 in number, possibly more, as the way in which their names are given in the directory allows for errors of Christian name, etc. Most operators were obviously one-man bands, others may have been small companies, such as Stanbury and Co. which operated a London-Dover service, or Pockett and Bates who ran a service between Canterbury and Deal. The pattern of these van services differs very much from that of the coaches. There are very few long-distance van services; London services start from the Medway towns or nearer still, apart from an odd service from Hythe. Van services outward are dominated by routes from Canterbury, 31 in all; only five places appear more than once as the destination, Dover [2], Goodnestone [2], Herne and Herne Bay [3] Margate [2] and Rochester [2], and the other destinations include tiny villages such as Hoath, Petham and Stowting, and which in 1841 had populations of 394, 646 and 276 respectively. In general terms, the starting points of the van routes are the larger towns of Kent, the destinations are the local large town or the local villages (see Tables 3 and 4). Most of the van routes TableABLE 3. NuMBER OF vAN SERviCES LEAviNG vARious PLACES in Kent, 1839 A ppledore 1 D eptford 2 Lydd 1 S heerness 1 A shford 8 D over 10 Maidstone 13 S ittingbourne 4 Broadstairs 1 F aversham 5 Malling 3 S taplehurst 4 Bromley 4 F olkestone 4 Margate 4 Tenterden 1 C anterbury 31 Gravesend 2 Mereworth 2 Tonbridge 3 C haring 2 H adlow 2 N ew Rom. 2 Tun. Wells 2 C hatham 6 H am Street 2 R ochester 5 Water’bury 3 C hevening 1 H awkhurst 2 S andgate 4 Westerham 2 C ranbrook 3 H erne Bay 1 S andwich 3 Wittersham 1 D artford 4 H . Halden 2 S eal 2 Wrotham 1 D eal 4 H ythe 5 S evenoaks 2 Wye 3 F.W.G. ANDRE WS 218 saw a daily return service (almost invariably excluding Sundays), a few three services in each direction a week, ten offered two services a day and there are a very small number [six] who ran a once-a-week service, such as the Canterbury-Hoath service referred to above, or the Canterbury-Nonington service. J ust what services the van operators offered is not explained. Obviously carriage of goods, but it seems certain that they offered carriage of passengers as well. Not all coaches were large; the largest seem to have carried four inside and eleven outside passengers, but the smallest carried but four, usually inside only: a van may well have offered seats to at least four passengers. In this context it is important to recall that walking what are by today’s standards long distances to and from work or market was perfectly usual;4 a walk of ten miles to market would be thought quite normal, so that the vans may well have acted as feeder services to the coaches, and their passengers may have set out on a considerable walk on alighting to get to their final destinations. Carriers The last group of road services is that offered by the carriers. These were certainly far more numerous than the van services or the coach services. The Pigot directory has 285 entries for services run, though many of these are clearly duplicates, in that they show parts of a much longer route, or the return working as a separate route; altogether it seems likely that approximately 218 separate carrier services can be identified, though some of them were duplicates, that is, two or more carriers operated over the same route. The actual number of carriers in business is in the TableABLE 4: NuMBER OF vAN SERviCES GOiNG TO vARious PLACES in Kent, 1839 (those over the border into Sussex asterisked) A shford 14 F aversham 3 N onington 1 S ittingb’rne 4 Barham 1 Good’stone 2 Petham 1 S towting 1 Brighton* 1 Gravesend 4 R amsgate 2 Tonbridge/ T. Wells 6 C anterbury 20 H astings* 7 R eculver 1 Westerham 3 C hallock 1 H erne Bay 3 R o’bridge* 1 Whitstable 2 C haring 1 H oath 1 R ochester 6 Wingham 1 C hatham 1 H ythe 2 R olvenden 1 Wye 1 C hislet 1 London 18 R omney 1 Y alding 2 C ranbrook 3 Lydd 2 R ye* 1 D eal 3 Maidstone 23 S andwich 1 D over 14 Margate 5 S heerness 1 THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 219 order of 180; though this may be too high a figure, owing to possible confusion between carriers with the same name whose Christian names are sometimes given and sometimes not, but the figure cannot be in error by more than 10 per cent at the most. The population of Kent in 1841 was just over 550,000, which suggests a ratio of one carrier to 3,000 of the population; if the 42 van operators are added to these, giving a total of 222 in the carrying trade, one to just under 2,500 persons. A ll these services may be assumed to be return services, even when this is not specified in the directory, and whereas at least half the van services ran a daily service, for carriers this was the exception rather than the rule. Most carriers offered a two or three times a week service, with only a minority offering daily journeys, and there are quite a number who offer only one service a week. Altogether these 218 carriers operated on 642 days between them, an average of almost exactly three times a week: the return journey was usually on the same day, but occasionally on the next, which would tend slightly to increase the total number of days during which carrier services were operating. Again, just what the carrier offered is not explained; at the time it must have been too obvious to need elucidation, but he must have offered carriage of goods to and from the place of destination and presumably from towns and villages en route. Either as a regular thing or by special arrangement he must have carried passengers – David Copperfield’s first encounter with Barkis was as a passenger on his cart – and as late as the 1920s the carrier who operated a service between Sandwich and Canterbury would take orders for things to be purchased in Canterbury and brought back for the customer.5 The carrier must have offered a much more personal service than the coach, and very possibly more so than the van.6 Everitt suggests that the carrier offered: - services as a shopping agent; - carriage for bulky goods to or from the nearest railhead, or town supplier; - public transport: Everitt suggests a maximum figure of a dozen, but if combined with a sizeable load of goods this implies a fairly large cart, almost a Conestoga wagon; - carriage of goods to market. The scale of these services, and their relative importance one to the other, can only be guessed. The number of towns and villages from which a carrier service started, or which one was advertised as passing through was 70, but of these the majority were home to one or two carriers only. As the county town, Maidstone led the way with 45 outward bound carrier routes, followed by Tenterden [14], Cranbrook [13] and Hawkhurst [9], just over a third of the total routes indicated. Many of the remaining 66 towns with an indicated F.W.G. ANDRE WS 220 service had only one or two routes. Apart from Maidstone, these towns with the numerous services are all in the south of the county, and they are noticeably poorly represented in Tables 3 and 4. Quite clearly the carriers were in the main providing a very local service, supplementary not only to the main-line coach services but to the van services as well. The towns (Table 5) to which the carriers travelled are, on the face of it dominated by London (85 cases) but this figure gives a false impression, as 49 of those cases are from towns west of the Medway, and broadly speaking on the line of the Dover road. The next most important destination was, as might be expected Maidstone (32 cases), but after that only Rye, with ten inward-bound services is in double figures; in other words the carriers were in the main going to the small towns and villages – and of course must have been serving the even smaller settlements which lay on the way. The average length of these carriers’ routes was in the order of 19 miles, a third of the routes covered less than ten miles, more than a third between ten and nineteen. The overall road picture of 1836-39 is therefore of one in which coach services – leaving aside the question of their cost – were only available on the main roads, where the service was more likely to be daily than not, but that this was very considerably supplemented by a van system, which tended to serve the towns rather than the villages, and an extensive carrier service which must have reached, directly or indirectly, almost every village in Kent. How far this can be thought of as providing a passenger, as opposed to a goods service, is a matter for debate. R oad transport in 1899 S ixty years later the overall picture had altered very considerably. By 1899 the railway system in Kent was almost complete, and the long-distance coaches had entirely vanished from the scene. Virtually all the van services had gone as well, but the carriers of Kent were very much alive and well. I n that year the directory lists 231 carriers in Kent, but this is certainly inaccurate. A further 57 carriers are listed in the town-by-town analysis, though their names do not appear in the classified list, and of those whose names are listed, 59 do not appear anywhere in the various town entries. It is possible that confusion has arisen in the compilation of the directory in that names have been wrongly heard, wrongly copied or just missed out, but it is evident that there must have been more than 230 carriers, perhaps as many as 280 or more, whether individuals or firms, large or small, operating in the county. A n analysis of the routes covered suggests that there were rather over 420 of these. Some of these are duplicates, in that two carriers operated a THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 221 service between A and B, and because there were two carriers, operating at different times or on different days, they have been counted as two routes. So far as possible return journeys have been eliminated. The situation often arises that Carrier A is shown as operating a service from TableABLE 5. PoiNTS OF STARTiNG AND ENDiNG ROuTES for the carriers of 1839 (those in Sussex asterisked) Place From To Place From To Place From To A ldington 1 F arnborough 1 Peckham E/W 1 A ppledore 2 1 F aversham 2 Penshurst 2 A shford 6 4 F olkestone 2 Pluckley 1 Bearstead 1 F rittenden 1 2 R amsgate 2 Benenden 1 Goudhurst 1 1 R o’bridge* 2 Bexley 2 Greenwich 6 R ochester 1 Biddenden 4 4 Groombridge 1 1 R olvenden 1 2 Blackheath 3 H adlow 2 R ye* 10 Boxley 1 H am Street 1 S evenoaks 6 1 Brasted 2 H astings* 2 S horeham 2 Bredgar 1 H awkhurst 9 S ittingbourne 3 Brenchley 2 H eadcorn 1 S marden 1 3 Brighton* 2 H igh Halden 1 S outhborough 1 Broadstairs 2 H ollingb’rne 1 St Mary Cray 1 Bromley 2 H orsmonden 1 1 S taplehurst 2 1 C anterbury 2 3 Keston 1 S utton Valence 1 1 C haring 3 1 Lamberhurst 3 S ydenham 2 C hatham 2 Lenham 1 Tenterden 14 4 C hislehurst 3 Lewes* 2 Teston 1 C owden 1 Lewisham 1 Tonbridge 2 1 C ranbrook 13 1 London 1 85 Tunbridge Wells 4 1 C rayford 2 Lydd 3 4 Wadhurst* 1 D artford 5 2 Maidstone 45 32 Warehorne 2 D eal 1 Malling 2 2 Wateringbury 1 D eptford 4 Marden 1 2 Westerham 1 D etling 1 Margate 1 Whitstable 2 D over 4 7 Mereworth 1 W. Wickham 1 D owne 1 Milton 1 Wittersham 1 E denbridge 2 N . Romney 3 1 Woodchurch 2 E gerton 1 N orth Cray 1 Woolwich 3 E ltham 2 N orthiam* 4 Wye 1 E whurst* 2 O rpington 1 Y alding 3 1 F.W.G. ANDRE WS 222 X to Y, and elsewhere in the directory he is shown as operating between Y and X, and at times and on days which make it clear that one trip was simply the return half of an out-and-back service; where these can be clearly identified only the single journey has been counted as a route. Other directory entries show what is obviously part of a longer trip, these too have been discounted where identifiable, but there remain the hard core of 424 as noted above. A part from the fact that there were very many more routes and carriers than there had been sixty years before, the most notable change is in the length of those trips. The average length of these journeys was now 11.0 miles, with a standard deviation of 7.9.7 If a list is made of the towns served (and it is not possible to present a ‘From’ and ‘To’ lists as it was in 1839, as it is in the majority of cases unclear in which direction the outward journey was undertaken) the overall picture is rather different to that of 1839. 178 towns appear as the start or end of a journey,8 but the lion’s share is taken by Maidstone (140 entries) followed by Ashford (55), Canterbury (52), Tunbridge Wells (38), Faversham (27), Tenterden (24), London (22) and Rochester (20); followed by Dover (14), Strood (13), Hawkhurst (12) and Rye (10). Between them these entries account for almost exactly half the towns served, all the rest, many being very small villages or hamlets, represented by single figure entries. Kent’s big towns (Maidstone, Ashford, Canterbury and Tunbridge Wells) account for a third of all towns served, Maidstone alone having over 16 per cent. There is, therefore, a much denser spider’s web of carrier routes effectively covering the whole county. By 1899 very few towns and villages in Kent were further than three miles from a railway station, and almost every hamlet was on somebody or other’s carrier route. The service, in many cases, was not exactly heroic; one trip on one day a week, but it certainly gave a far better opportunity for country folk to reach a town themselves, or for them to place commissions with a carrier who was operating over a much shorter distance, with all the advantages of local knowledge which that implied. The extent of this coverage, just considering the services to or from Maidstone, is made very clear by Map 1. Each town or village whose name appears is listed as having a carrier service in 1899; so many per day or per week – ‘1D, 3W’ means that there was a daily service in addition to three weekly services, for example. Clearly towns and villages nearest to Maidstone will have benefited from services to more distant settlements. Thus, Loose (just south of Maidstone and today a suburb of the town) will have been served by all the carriers going to Goudhurst, Staplehurst, Cranbrook and so on; there must have been a steady stream of carriers’ carts passing though the village, more or less all day. A similar (though not so intensive) pattern exits round any of those settlements with a multiple entry of carriers’ services, as listed above. THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 223 J ust what effect these increases in the availability of road transport had on the villages of Kent is much more difficult, if not impossible, to quantify. Research on the effect of the coming of railways on east Kent has suggested that while the ‘suburbs’ of the larger towns suffered from the competition for goods and services which those towns were able to offer, the villages, both on and off the railway, were not so badly affected, and may even indicate a minor growth in prosperity.9 The evidence is however thin – the main source of information, the directories, do not have many entries for most villages and so any statistical analysis must Map 1. Maidstone Carrier Service Routes 1899. F.W.G. ANDRE WS 224 be viewed with caution. Certainly the village shopkeeper will have found it easier to obtain supplies from the local big town, via the local carrier, which the big town was in turn able to obtain via the railway. A further point of interest is in the total distance the various carriers covered. In 1839 all the carriers’ journeys for a week totalled 25,868 miles; by 1899 that figure had risen to 31,176, an increase of 20 per cent. The lengths of the various carriers’ trips had changed very considerably also (Table 6). By 1899, well over half of all the carriers’ journeys were less than ten miles in length; in 1839 a similar percentage had been up to 15 miles long. Many were very short indeed in 1899; 36 per cent were less than 7.5 miles long; in 1839 only about 20 per cent had been so short. A ncillary needs A very considerable capital investment must have been involved. Each carrier must have had at least one horse and at least one cart and though in some cases in the carrier services, taken at random from the 1899 carriers’ list, such as Wheeler, who set out from Maidstone on Thursdays at 1.00pm on his way back to Charing, via Park Gate, Sandway and Lenham on what appears to have been his only advertised carrier’s work of the week – there must have been other work for the horse and perhaps the cart on the other days of the week. Where there was a daily journey – e.g. Williams’ daily trip over the 18 miles from Hawkhurst to Maidstone, via Sissinghurst and Loose in that same year – there may well have had to be more than one horse and more than one cart involved. Thirty-six miles in a day for a horse would have been a very long journey, much TableABLE 6. COMPARiSON OF THE LENGTH OF CARRiERS’ jouRNEys, 1839 and 1899 Distance (miles) 1839 1899 N umber Per cent N umber Per cent 0 - 2.5 2 0.9 6 1.4 2.6 - 5.0 19 8.5 54 12.7 5.1 - 7.5 25 11.2 95 22.3 7.6 - 10.0 33 14.8 87 20.4 10.1 - 15.0 47 21.1 108 25.3 15.1 - 20.0 27 12.1 44 10.3 20.1 - 25.0 17 7.6 14 3.3 25.1 - 30.0 10 4.5 7 1.6 30.1 - 40.0 19 8.5 5 1.8 40.1 - 50.0 11 4.9 3 0.7 O ver 50.1 13 5.8 3 0.7 THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 225 longer than a stage-coach ‘stage’ had been, and one horse cannot have been expected to do the trip day after day, or even the whole trip by himself.10 A coaching stage seems to have been about ten or twelve miles, so Williams may have had to have three horses for the trip, one for the first twelve miles, one to cover the last six into Maidstone and back, and a third horse to take him and the cart back to Hawkhurst, all of which will have cost extra in livery. H orses cost money: London omnibus horses cost about £20 each in 183011 and though a country carrier’s horse is likely to have been at the bottom end of the market, this figure may be accepted as a basis for discussion. Costs certainly rose during the century: Wordie and Co. of Glasgow were paying an average of £28 2s. in 1858, and by 1879 the average cost had risen to £58.12 Horses bought by the Rochdale Canal Company came into the same broad range of prices; £40 in 1889, though a young (and probably untrained) horse was only £30.13 Old, and presumably worn-out horses were, of course, cheaper: the Isle of Wight Central Railway Company sold a horse, complete with his harness and the cart itself to Pickford’s for £30 in 1904.14 Food had to be found too – 15s. a week is suggested15 – as well as the cost of shoeing and the occasional farrier’s bill. These sums were reckoned to total £54 15s. a year (just over 21s. a week) by the Rochdale Canal Company.16 Wordie’s were paying 7s. 7d. a day at Stirling in 1868, and 6s. 8d. at Stranraer and Stonehaven.17 With an agricultural labourer’s wages at only about £1 a week, a horse, his equipment and his upkeep represented a sizeable investment: the village carrier can hardly have been a man of financial straw. These horse costs seem to have been at the upper end of the scale. In the early years of the twentieth century Henry Spicer ran a small greengrocery business in Sandwich, and in July 1911 he bought a horse for £3 15s.18 He obviously also had at least one other horse, a cob, as in January 1912 he paid 2s. 8d. for four shoes for it,19 and three days later had to buy two shoes for the horse, which cost him 3s.20 In June 1912 he bought a pony for £3, with a second-hand set of harness thrown in for 7s. 6d.;21 it is not clear from the records if the pony replaced the horse and/or the cob. There is some indication from his records that he may have been involved in a carter’s business in the town as well. He rented a stable in the town for 6s. 6d. a month22 and between April and May 1912 he had to buy £1 15s. worth of hay, straw and fodder, presumably for the horse or horses.23 There is an indication throughout all the receipts that Spicer was only staying in business by the tips of his financial fingers, but between them the animals must have cost some £15 a year in feed, and £3 18s. to stable, plus shoeing, etc., a total surely in excess of £20 a year, a fairly considerable drain on his financial resources.24 E ach horse must have been stabled or at least fed and watered at the end F.W.G. ANDRE WS 226 of the outward trip and some inns clearly specialised in this work. In 1899 at Faversham it was the Ship (10 routes), at Tunbridge Wells the Hand and Sceptre (7 routes), at Strood the Angel (7). At Maidstone three houses shared the volume of traffic, the Queen’s Head (16 routes), the Rose and Crown (18), and the Star (24). The smallest carriers were often advertised as starting from ‘His house’, but that can only have been half the story. Intermediate stops for all carriers were almost certainly going to be the local inn, which brought that inn more trade; the range of spin-off benefits from the carriers’ services must have been very considerable with so many carriers, inns, farriers, blacksmiths, wheelwrights and so forth involved. A similar picture exists for 1839, though the names of the inns are different – in Maidstone the Castle Inn, the Haunch of Venison, the Queen’s Head, the Ship and the Sun were the most popular places; in Faversham it had been the Coal Exchange and at Dover the Packet Boat. I n 1839 the names of the coaching inns are indicated in four cases out of five, but the most interesting thing about these names is that in very few cases do they replicate the carriers’ inn names of that year, or the carriers’ inns of 1899. The Haunch of Venison and the Queen’s Head are mentioned at Maidstone, but only once each, and the favoured coaching inns at Tonbridge were the Bull and the Rose and Crown, and at Dover the Victoria and Gun [sic].This may be because the name of the inn has changed, or it may represent the result of the complete change in the nature of the traffic from the essentially passenger traffic of the coaches, to the primarily goods function of the carriers; different inns offering different (and presumably more convenient) facilities moved in to take up the new trade. All the same, it is surprising that there is not a greater degree of continuity between the 1839 and the 1899 carriers’ town bases; it might have been thought that the facilities needed in 1899 were those which had been required sixty years ago – unless the increasing scale of the carriers’ coverage made more commodious premises necessary. C hange of pattern of services The point has already been made that the 1839 carriers’ average journey length was about 19 miles, though two-thirds of them were shorter than that, and that by 1899 the length of the average service had come down to 11 miles. In 1839 the 180 carriers in business actually made 699 trips a week between them, since some offered a daily journey between the towns they served and those journeys covered just under 26,000 miles each week, since each carrier had to return to his base after each journey. By 1899 the number of carriers had risen to at least 230; between them they ran about 1,200 journeys each week, covering a grand total of over 32,500 miles. Thus, though the number of journeys had not quite doubled in the sixty years involved, the actual mileage covered had increased THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 227 by fractionally under a quarter, emphasising the frequent, short journey nature of the 1899 carriers’ trade compared to that of his forebear. This point is made more evident by a comparison of the average length of the carriers’ routes in 1839 and in 1899 from the main centres of Ashford, Canterbury and Maidstone, together with the standard deviations of those averages (Table 7). TableABLE 7. AvERAGE LENGTHS in miLES OF CARRiERS’ jouRNEys fromFROM ASHFORD, CANTERBury AND MaiDSTONE in 1839 AND in 1899, showSHOWiNG THE STANDARD DEviATiONS FROM THOSE averages Town 1839 1899 A shford J ourneys 10 55 Average (miles) 19.5 12.8 S td deviation 14.2 5.96 C anterbury J ourneys 5 63 Average (miles) 20.4 10.91 S td deviation 15.04 4.29 Maidstone J ourneys 77 145 Average (miles) 14.1 10.78 S td deviation 5.76 4.75 I n 1839 there were only ten identifiable carriers’ routes from Ashford, then a small market town, though having a large hinterland of the Romney Marsh as its sphere of influence, and these were operated by eight carriers. Two of these routes served London, artificially inflating the average length of the routes, as indicated by the high standard deviation. By 1899 there were 55 routes into or out of Ashford, but their spread was evidently very much more limited, having an average journey of just over twelve miles, and a standard deviation rather under half that of 1839. I n 1839 there were only five carriers’ routes which started from or ended at Canterbury, but as one of these served London, the average length is quite high, with a deviation of well over half the average distance (15.04 against 20.4): by 1899 there were 63 routes starting from or ending in the city, two of which fed London (strictly, they were advertised as to ‘All parts of the kingdom’) but the rest of which were very local; the average length is less than Ashford’s, and so is the standard deviation. Maidstone, of course, was the county town, and even in 1839 there were 77 identifiable routes which began or ended there, and though they effectively covered most of the county, the average length of the journeys was only 14.1 miles, indicating the extent to which the Maidstone locality received the lion’s share of the services, a point underlined by the low SDsd F.W.G. ANDRE WS 228 of 5.76. By 1899 there were 145 services, with the low average of 10.78 miles and the low SDsd of 4.75: London still had a couple of services, but in common with the other two towns cited, the long journeys had obviously been taken over by the railway. Taking into account the fact that by 1899 very few places in Kent were more than three miles from a railway station, it is very clear that there was an ample service of public transport covering the whole county. Hardly anyone can, by 1899, have lived more than, say, a mile from a station or a recognised carrier service, and for most people it must have been less. Certainly the rumours so prevalent in the early years of the railway of the imminent disappearance of the horse as a means of transport were obviously greatly exaggerated. O ther road transport By 1899 there was a new development in the form of the urban tramway. The way had been eased for the construction of such lines by the Tramways Act of 1870, but though a horse tramway had been proposed in 1873 to fill the gap between Dover and Deal and to continue thence to Margate via Deal, Sandwich and Ramsgate, nothing was ever done, and the Hythe and Dymchurch Tramway (1892-93) and Lydd Tramway schemes (c.1880) appear to have been equally abortive. The first tramway to operate in Kent was the horse-powered Folkestone, Hythe and Sandgate tramway, which opened in 1891, with a total length of 3.36 miles: apart from the Herne Bay Pier Tramway of 1899 the only other tramway to be opened in Kent during the nineteenth century was the 1897 Dover electric line. This was originally only three miles long, though a mile and a quarter extension was added in 1905.25 On the original line a five-minute headway was claimed through the town, and a ten-minute on the road to the Harbour, and this must have been justified by the results; in the first year 1,794,905 passengers were carried at a flat fare of 1d., nearly 5,000 a day. The horse omnibus proprietors had charged 4d. for a similar journey, clearly accounting for the new trams’ popularity.26 By the end of the century, yet another form of road transport had appeared in Kent: this was the omnibus. The 1882 Kelly’s Directory of Kent lists ten omnibus proprietors, by 1899 there were 18: towns where these were centred are listed in Table 8. This list is a very interesting one. Cranbrook and Tenterden were the only towns without a railway station, though in 1895 and 1896 Acts were passed which eventually connected Tenterden with the South Eastern railway at Robertsbridge, and by 1899 the new line was well advanced – it was to open in April, 1900. The other towns tended to be bunched together – Beckenham and Sidcup; Chatham, New Brompton THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 229 TableABLE 8. NuMBER OF KENTiSH OMNibus PROPRiETORS (in townsTOWNS WHERE BASED) in 1882 AND 1899 1882 1899 1882 1899 A shford 1 Paddock Wood 1 Beckenham 1 2 R ochester 1* C hatham 2 (1*) S evenoaks 1 C ranbrook 1 S idcup 1 D over 2 1 S ittingbourne 1 H erne Bay 1 1 S trood 1* H ythe 1 1 Tenterden 1 Margate 1 Tonbridge 3 N ew Brompton 1* Tunbridge Wells 2 2 * One Company, John Bean Martin, had depots at each of these places, but has been shown in the table as if it was four companies. (Gillingham), Rochester and Strood form the Medway Towns conurbation; and Tonbridge and Tunbridge Wells are cheek by jowl. Between them these three groups comprise 13 of the 20 proprietors’ establishments in 1899. Apart from these, the omnibus proprietors were well scattered. More interesting than the towns which had an omnibus service are the towns which did not. Maidstone, Kent’s county town was apparently not so served, nor was Canterbury which was in many ways the regional capital of east Kent. Thanet only had a single operator. If these services did not exist, it seems reasonable to assume that they were not regarded as potentially economic: Maidstone after all was very well served by a network of carriers’ carts, as (to a lesser extent) was Canterbury. Shanks’ pony must have been in regular use in Thanet, and the presence of the two competing railways (South Eastern and London, Chatham and Dover) meant that there was a very intensive almost suburban service connecting the three stations in Margate with the three in Ramsgate and the singleton at Broadstairs,27 which presumably meant that there was no economic raison d’être for an omnibus service there. I nformation on horse bus routes and fares seems very hard to come by, except for London.28 In the 1892 edition of the Official Guide to the South Eastern Railway it was noted that at a number of stations, ‘Omnibuses meet every train’, but no indication was given as to places served, suggesting a simple link between station and town. These stations were at: Ashford, Canterbury, Dover, Maidstone, Margate, Ramsgate, Sevenoaks and Strood. This list is at considerable variance with that derived from the entries in the 1899 Kelly’s Directory; in only three cases is there correspondence (Dover, Sevenoaks and Strood). F.W.G. ANDRE WS 230 I n three other cases a route was indicated, omnibuses from Dartford serving Farningham and Sutton at Hone; from Appledore serving Tenter-den (described as a coach, twice a day) and from Etchingham serving Hawkhurst. Hawkhurst, of course, did not have its own station (and that very remote indeed from the town centre) until 1893, but since Farning-ham and Sutton at Hone had had a station since 1860, that omnibus was presumably for the benefit of the settlements which lay between them and Dartford. No omnibus proprietors are indicated at any of these three towns in the 1899 Kelly: these discrepancies may indicate a major change in transport patterns over the seven years involved, or it may be that the omnibus proprietors were considered to be too small commercial fry to be worthy of a directory entry. Two very dedicated services were advertised: the Black Horse Hotel at Deal sent an omnibus to meet every train – though the station is little more than four hundred yards distant – and the Granville Arms Hotel at St Margaret’s Bay boasted that a ‘well-appointed omnibus plies between the Hotel and Martin Mill station’, but these two services were obviously available for hotel patrons only. By 1907, when the South Eastern and Chatham Railway published the sixth edition of its Official Guide, even these modest references had vanished, possibly because with two hitherto competing railways’ coverage to be included, space was at something of a premium. The railway itself advertised that Messrs Charles Rickards, Ltd, was ready to supply ‘Private omnibuses for Family Parties’ which could hold ‘six persons inside, and one outside’ to carry passengers to or from the various London stations, but these cost at least 1s. per mile, with a minimum charge of 4s. In Canterbury the County Hotel omnibus met all trains at the East station (the LCDR station) and trains at the West station (SER) by appointment. The only link between the Guide information and the 1899 Kelly’s Directory list is provided by Daniel Hoadley, who had been in business at Tunbridge Wells certainly as early as 1882, but no indication is given as to whether or not he provided what might be recognised as a ‘bus service’ today. Since the order office was given as being at the South Eastern railway station gates, it seems possible that he did no more than provide a service between the stations and various parts of the town, on an ‘as required’ basis. S o far as can be ascertained, this seems to have been the normal pattern of omnibus services in the last years of the nineteenth century in Kent, connecting the local station with the centre of population, but information is very patchy. In 1874 there were three concerns running omnibuses between Hythe, Sandgate (the branch railway to which had just been opened) and Folkestone. Ovenden ran three buses daily to Folkestone in addition to meeting every train.29 Two other operators also worked the THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 231 F olkestone service, each making three trips. This service must have been a popular one, as by 1892 each of the last two operators had increased their daily journeys to six trips daily.30 The 1905 edition of Kelly’s Directory of Kent lists three Motor Omnibus Proprietors, one offering a service between Dartford and Farningham, and another a service between Headcorn and Maidstone, via Sutton Valence; the third was at Herne Bay, but no route is indicated, nor any frequency suggested for the other services. The two indicated services do seem to represent a change of pattern; Dartford and Farningham were both served by the railway (even if Farningham Road was some distance from the village), as were Maidstone and Headcorn; these routes are perhaps the beginning of the expansion of the motor bus services which in the course of decades were to put the rural railway out of business. C ars and lorries were only appearing in small numbers at the end of the century. The 1899 edition of Kelly’s Directory of Kent mentions two ‘Motor Car Makers’, one at Tonbridge and the other at Ashford, as well as a single ‘Motor Wagon Maker’ at Maidstone, but only six years later, in 1905 the Directory lists nearly two columns of traders selling or supplying motor cars, car parts and garaging services. Fourteen of these entries were for businesses which claimed to be car makers or builders, but to judge from information given about chassis numbers of the various motor manufacturers at work in England in 1918,31 production in Kent must have been on a minute scale, and the effect of the internal combustion engine on the transport scene minimal. Wind in the Willows appeared in 1908: the appearance of the first motor car must have had much the same effect on of the inhabitants of Kent that it did on Toad. V olume of traffic on roads No figures seem to exist for how much traffic actually travelled along the roads of Kent, apart from the calculations made as to overall volume of traffic which the railways expected (or hoped) to take over from road services. The best one can do is make a series of educated guesses, based on what little material is to hand. F igures survive for the rates at which the gates on the Sandwich-Dover turnpike were leased out for the years 1833-1873 (with certain gaps);32 in 1839 the Trustees, whose financial position was at that time very precarious secured a private Act which laid down a new table of tolls to be charged at the various gates along the road.33 Halfway along this road lay the gate at Upper Deal; in 1844 this was leased out for the sum of £452, which was in fact the highest figure ever secured. Assuming that the lessee expected to make a profit on the deal, the gross income must have been more than this, possibly between 150% and 200% of that F.W.G. ANDRE WS 232 figure. Assume for the purposes of argument that the income was 150% the rental, say £678 per year. The Act laid down the new charges for each gate, and though this list is very detailed it will perhaps be enough for the present purposes if two classes of toll-payers are considered, the first those who were passing through the gate with a passenger vehicle of some sort, or with a cart with narrow wheels: these two groups both paid 3d. per passage of the gate, though this was a ticket valid for the whole day; a second passage would not incur a second charge. If all the traffic through the gate consisted of this sort of class (which it obviously would not have done), at 3d. a time the estimated gross income at the Upper Deal gate would imply over 54,000 vehicles passing through in the course of a year, 148 a day. Other toll-payers in general, the second class, paid less per passage – single horses only 1d., wagons with wider wheels 2½d. or 2d. according to the wheel width – but droves of animals were more expensive; large animals 10d. per score, small animals 5d. per score. Again assuming that all the traffic through the gate consisted of droves of cattle this would suggest that 44 score per day passed through. There is the further point that the Act laid down that the Trustees could decree that a toll paid at Gate A could open specified other gates along the road at their discretion, so the figures calculated above must be minimum figures for actual passage through the Upper Deal gate. Obviously the traffic passing the gate must have been a mixture of all classes and types, and it is impossible to guess just how the traffic was comprised. All the same, the suggestion is that if only the most expensive traffic (passenger traffic and carts) is considered there must have been something like 12 units passing in each of the twelve hours of average daylight; the inclusion of cheaper units will increase this hourly rate considerably, and if the lessee took more than 150 per cent of his rent in tolls, the figure rises again. Twelve an hour is not exactly M25 level of traffic, but it is more than might have been expected on a relatively minor road in the extreme east of Kent. C osts of road transportation I nformation on this matter simply seems not to exist, apart from the odd reference such as that to the price of Dover horse buses, referred to above. None of the carriers seem to have advertised their services in the local press in east Kent, and there is no indication of how much was charged. Presumably fares, or costs, were based on what the traffic might be expected to bear, but just what those charges were seems to be impossible to discover. Clearly they must represent the cost of the horse and his harness, his feed and stabling, his shoes and the odd farriers’ bill, which might come to about 15s. or more a week per horse (see above) plus the carriers’ own time and profit, plus any contribution to some sort THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 233 of depreciation account against the day when a new probably second- or third-hand horse was needed.34 Equally, if third-class railway fares were of the order of 1d. a mile, the carrier cannot have charged a great deal more after taking into account the convenience factor that his more-or-less door to door service could offer. In 1859 the Manchester, Sheffield and Leicester railway agreed with Thompson McKay to carry ‘smalls’ (i.e., under 1 ton) for 3d. per consignment,35 and in 1896 the Lancashire, Derbyshire and East Coast Railway contracted with Pickford’s to carry and deliver parcels for 1½d. each; heavy goods were 1s. 6d. per ton from station to station; if collected an delivered, 2s. 6d. per ton.36 As late as 1926 Wordie’s were charging only 2d. for ‘smalls’, collected and delivered.37 SummarMMARy The long-distance road services of the pre-railway era did not survive the railway’s arrival, but local road services certainly increased in intensity though their routes were probably considerably shorter. By 1914 very few villages were more than three miles from a railway station, and that gap had been filled, directly or indirectly, by road transport. The commercial and social value of feeder networks to the nearest convenient railhead is implicit in the construction of various (almost invariably unsuccessful) light railways in the period 1890-1914. The rapid development of the internal combustion engine during the First World War and the appearance on the roads after 1919 of fleets of ex-WD lorries, picked up for a song by those who had learnt to drive them in the war killed off those railways, and must have done great harm to the horse carrier. Mr Williams, who might have had to have three horses to run his Hawkhurst-Maidstone service, would only need one lorry to offer the same, or a faster, service. What the railways failed to do to the horse, the lorry and the internal combustion engine did between them. endnotes 1 Pigot and Co.’s Royal National and Commercial Directory and Topography of the counties of Kent, Surrey and Sussex (London, 1839). 2 Kelly’s Directory of Kent (London, 1899). 3 Bates, A., Directory of stage coach services, 1836 (Newton Abbott, 1969). T.P. Smith’s article ‘The Geographical Pattern of Coaching Services in Kent in 1836’, Archaeologia Cantiana, xcviii (1982), 191-213, provides a very comprehensive analysis of Bates’ material for Kent. 4 Bagwell instances a Gloucestershire carpenter who in the 1880s regularly covered 28 miles a day, six days a week, to and from his daily work. ‘The decline of rural isolation’, P.S. Bagwell, in G.E. Mingay (ed.), The Victorian countryside (London, 1981), p. 32. 5 Oral history evidence: Mr Joe Trussler. F.W.G. ANDRE WS 234 6 The role and position of the country carrier in rural society at the end of the nineteenth century is exhaustively considered in Everitt, A., Landscape and community in England (1985), pp. 279-307, Chapter 11, ‘Country carriers in the nineteenth century’, which extends his earlier work, ‘Town and country in Victorian Leicestershire: the role of the Village Carrier’, in Everitt, A. (ed.), Perspectives in English Urban History (1973). 7 Route lengths have been calculated using Microsoft’s Autoroute computer programme. Motorways have of course not been used, and the route chosen has been that which seems to have been most likely, given the information supplied. Any errors are likely to be roughly self-cancelling. 8 A again, the information supplied in the directory means that this can only be an approximate figure; some routes may have begun or ended at one of two or even three small villages. 9 S see Andrews, F.W.G., ‘The effect of the coming of the railway on the towns and villages of East Kent, 1830-1914’ (unpublished Ph.D thesis, University of Kent at Canterbury, 1993), Ch. 12: ‘Railways and the Community: the Kentish Evidence’, Archaeologia Cantiana, cxxiii (2003), 185-202. 10 S some idea of the complexity of running a carrier’s business can be gained from E. Paget-Tomlinson’s, The Railway Carriers (Lavenham, 1990) which deals with the organisation of Wordie and Co. of Glasgow as late as the 1930s. This was of course a very large company indeed, which specialised in carrier work for the London, Midland and Scottish Railway and which had well over 2,000 horses on strength. 11 T.C. Barker, ‘Urban Transport’, in Freeman, M.J. and Aldcroft, D.H., Transport in Victorian Britain (Manchester, 1988) p. 139. 12 E. Paget-Tomlinson, op. cit. (see note 10), p. 49. 13 R richard Dean, ‘Three hundred and seventy two horse power’, in Journal of the Railway and Canal Historical Society, No. 184 (March, 2003), 238. 14 R.J. Maycock and R. Silsbury, The Isle of Wight Central Railway (Usk, 2001), p. 110. 15 T.C. Barker, op cit. (see note 11). 16 R r. Dean, op. cit. (see note 13). 17 E. Paget-Tomlinson, op. cit. ����������������������(seenote10),p. 106.(see note 10), 106. 18 S sandwich Archives; Spicer receipts; item 217. 19 Ibid., item 247. 20 Ibid., item 243. 21 Ibid., item 194. 22 Ibid., items 77, 95, 118, 130, 141, 171 and 199. 23 Ibid., items 146 and 166-68. 24 Thompson suggests that a town horse cost in the order of 10s. a week to feed: if this is the case for Spicer’s horses, he must have had access to feed other than that which is represented in his accounts, which work out at only about 5s. a week. Thompson, F.M.L., ‘Nineteenth-Century Horse Sense’, in Economic History Review, Second series Vol. XXIX, 78. 25 I information on the Folkestone, Herne Bay and Dover tramways from Turner, K., The Directory of British Tramways (Sparkford, 1996). 26 Horn, J.V., The story of the Dover Corporation Tramways, 1897-1936 (Cricklewood, 1955), Ch. IIii. According to Sherrington, C.E.R., 100 years if Inland Transport (1969), p. 274, the average bus fare in London was well under 1d. per passenger mile, with an overall range of ¾d. to 1¼d. If the horse-bus drivers of Dover charged 4d. for a trip of less than two miles, this was well over twice the most expensive London rate. 27 Margate West (LCDR), Margate Sands (SER), Margate East (LCDR), Ramsgate Town (SER), Ramsgate Harbour (LCDR), St. Lawrence (SER) and Broadstairs (LCDR). THE ROAD SERVICESservices OF KEN T IN THE NINE TEEN TH CEN Tury 235 28 See Barker, T.C. and Robbins, M., A History of London Transport; Volume I: The Nineteenth Century (1975), Ch. VIvi, passim. 29 When the line opened in October 1874 there were three trains daily from, and five services to, Ashford, but there was also a shuttle service between Hythe and Sandgate, with five additional up services and six down. Hart, B., The Hythe and Sandgate Railway (Didcot, 1987), p. 15. Presumably only the Ashford connections to Hythe are meant here. 30 H hart, B, op. cit. (see note 29), p. 125. 31 Motor Car Index, 1918-1929 (Norwich, 1929, reprinted Brighton 1961), passim. 32 A andrews, F.W.G., ‘The Sandwich to Dover turnpike, 1833-1874’, Archaeologia Cantiana, cxvii (1997), 10. 33 An Act for repairing the road from Dover in the county of Kent through Deal to Sandwich in the said county; Private Acts, 2 Vict. c. xxxiii. (1839). 34 Wordie’s calculated that in the nineteenth century horses just, but only just, earned their keep. By 1925 however, horse traction brought in £394,501 against a cost of £293,777, some 34% profit. E. Paget-Tomlinson, op. cit. (see note 10), p. 106. 35 D dow, G., Great Central (Vol. I), p. 194. 36 D dow, G., Great Central (Vol. IIIiii), p 167. 37 Paget-Thompson, E., op. cit. (see note 10), p. 95.