256 POPULAR SCIENCE MONTHLY. 



ent cities of Genoa, Nice, and Marseilles. There it divided: one branch 

 turned north directly through France, passing by many important 

 cities till it reached a point near Boulogne, where, as at Brundisium, 

 boats were always in readiness to carry passengers to Britain; the 

 other branch, starting from Marseilles, traversed the Spanish Penin- 

 sula to its remotest part. England had its own roads ; beginning near 

 Dover, a leading artery ran to the north almost straight to York, 

 while the others traversed the island from east to west. 



Tlie length of the Roman system of roads from the Wall of 

 Antoninus in Britain to Jerusalem was four thousand and eighty 

 miles, and, to the extreme limits of the empire on the Euphrates, 

 about four thousand five hundred miles. From city to city the 

 Roman roads usually went in a straight line, property being con- 

 demned and appropriated for the public use without the slightest 

 regard for the feelings or rights of its owners, while natural obstacles 

 were almost ignored. Mountains were tunneled, morasses filled with 

 stones and* earth ; up one side of a hill and down the other went the 

 road, for, as travel was altogether on foot or on horseback, and 

 wheeled vehicles were not used in the country, a steep grade was no 

 objection. Bold arches of heavy, solid stones were thrown across 

 the smaller streams, while great bridges spanned the rivers. Trajan's 

 bridge over the Danube had twenty-one piers of stone built on 

 piling; each end was fortified by a camp and outworks, and when 

 the structure was destroyed its ruins blocked the river. Many 

 Roman arches over mountain gorges and smaller streams are still 

 in existence. In "Wales, the Devil's Bridge near Aberystwyth bears 

 testimony to the solid and lasting character of the work done by 

 Roman engineers; in Spain and Portugal many Roman arches 

 remain to show that honest building could be done in the days of the 

 Caesars. Here and there in the Alps, in the Pyi'enees, in the Tyrol, 

 in the Balkans, the remains of a Roman road lead up to the side of a 

 mountain where a half-closed or entirely blocked tunnel once pierced 

 the giant mass. Here and there in the lowlands of the Danube, in 

 the plains of the Po, on the lower waters of Belgium, the trace of a 

 highway leads into the depths of now impassable quagmires, which, 

 however, were no obstacles to the indefatigable road builders. 



They did their work well. When a route had been surveyed and 

 a line of road from eight to twenty feet wide was staked out, the 

 surface of the earth within the inclosed limits was all removed until 

 the clay bed was reached. This was densely packed with rammers: 

 a layer of small stones was placed in position and rammed into the 

 clay with heavy mallets, then another layer of stones and sand with 

 cement; then a layer of sand or gravel with lime mortar; then layer 

 after layer of broken stones cemented into one mass, and above all 



