It is important to insist upon this point, because there is nothing commoner than the statement that our prehistoric roads commonly follow the very tops of the hills. The reason usually given for this statement is, that the tops of the hills were the safest places. One could not be ambushed or rushed from above. But the condition of fear is not the only state of mind in which men live, nor does military necessity, when it arises, force a laden caravan to the labour of reaching and following a high crest. The difficulty can be met by a flanking party following the top of the ridge whose lower slope is the platform for the regular road. In this way the advantage of security is combined with the equally obvious advantage of not taking more trouble than you are compelled to take. There are several excellent examples of the flanking road on this way to Canterbury, notably the ridge-way along the Hog's Back,[6] which has become the modern high-road; but nowhere does the Old Road itself climb to the top of a hill save when it has one of two very obvious reasons: (a) the avoidance of a slope too steep to bear the traveller with comfort, or (b) the avoidance of the ins and outs presented by a number of projecting ridges. On this account at Colley Hill, before Reigate, and later, above Bletchingly, near Redhill, the Road does climb to the crest: and so it does beyond Boughton Aluph to avoid a very steep ravine. Once on the crest, it will remain there sometimes for a mile or so, especially if by so doing it can take advantage of a descending spur later on: as after Godmersham Park down to Chilham. But, take the Road as a whole, its habit when a convenient southerly bank of hills is present, is to go up some part of the slope only, and there to run along from 50 to 100 feet above the floor of the valley. If it be asked why the earlier traveller should have been at the pains of rising even so much as this, it may be noted that the reason was threefold: it gave the advantage of a view showing what was before one; it put one on the better drained slope where the land would be drier; and it lifted one above the margin of cultivation in later times when men had learned to plough the land. In the particular case of this Road it had the further advantage that it usually put one on the porous chalk and avoided the difficult clay of the lower levels.

IV. Wherever the Road goes right up to the site of a church it passes upon the southern side of that site.

It is necessary to digress here for a moment upon the archæological importance of these sites. They are but an indication, not a proof, of prehistoric sanctity. It would be impossible to say in what proportion the old churches of these islands stand upon spots of immemorial reverence, but it is certain that a sufficient proportion of them do so stand (the church of Bishopstoke, for instance, upon the site of a Druidical circle), as to afford, when a considerable number are in question, a fair presumption that many of the sites have maintained their meaning from an age long prior to Christianity. Apart from the mass of positive evidence upon this matter, we have our general knowledge upon the methods by which the Faith supplemented, and in part supplanted, older and worse rituals; and there is an impression, which no one who has travelled widely in western Europe will deny, that the church of a place has commonly something about it of the central, the unique, or the isolated in position; characters which cannot wholly be accounted for by the subsequent growth of the community around them.

Now, on its way from Winchester to Canterbury, the Old Road passes, not in the mere proximity of, but right up against, thirteen existing or ruined churches. They are, proceeding from west to east, as follows: King's Worthy, Itchen Stoke, Bishop Sutton, Seale, Puttenham, St. Martha's, Shere, Merstham, Titsey, Snodland, Burham, Boughton Aluph, and Chilham.[7] In the case of eight it passes right up against the south porch; in the case of two (Bishop Sutton and Seale) it is compelled to miss them by a few yards. One (St. Martha's) is passed on both sides by a reduplication of the track. One (Chilham) is conjectural, and the last (Shere) is doubtful.

THE CHURCH OF SHERE

The habit is the more remarkable from the fact that the Road commonly goes north of a village, and therefore should, unless it had some purpose, commonly go north of all churches. And, indeed, it does pass many churches to the north, but it always leaves them (as at Chevening, Lenham, Charing, and the rest) to one side. It never goes close to their site.

The importance of this rule will be apparent when we consider, later on, the spots in which a church stands, or has stood, and where, at the same time, the track is doubtful and has to be determined.

V. In crossing a river-valley, the Road makes invariably for the point where spurs of dry ground and rising ground come closest upon either side, and leave the narrowest gap of marshy land between.