392 THE POPULAR SCIENCE MONTHLY. 



the division of the population having been ascertained, how shall we 

 lay out the roads so as to make the different lines of communication, 

 particularly the most frequented ones, as short as possible? Four 

 plans ordinarily present themselves under which the ground selected 

 for our city to be built on may be laid out : the square, hexagonal, oc- 

 tagonal, and circular. Of these, the hexagonal type seems, according 

 to our mathematical calculations, the one which gives the greatest 

 length of streets and the greatest amount of habitable surface, with 

 the smallest consumption of space. As a rule, for cities the population 

 of which is homogeneous, the octagonal and hexagonal plans are much 

 preferable to the square in respect both to the utilization of the sur- 

 face and the facility of communications. Unfortunately, both plans 

 are liable to the objection that they give house-lots having acute an- 

 gles of sixty degrees in the hexagonal, and of forty-five degrees in the 

 octagonal plan. Practically the population finds it convenient to bear 

 toward the center of the city, and, the more it bears that way, the more 

 it is to its interest to do so. Consequently, the density of the popu- 

 lation diminishes from the center toward the circumference. Under 

 such conditions, the circular plan is very satisfactory. It has in effect 

 the double advantage of furnishing direct roads to the center of the 

 town, and of accommodating itself to giving to the houses in the out- 

 lying quarters the greater amount of space they require. For the 

 center of cities, where the population is compact and homogeneous, 

 where land is dear and communication is needed in every direction 

 alike, the hexagonal plan, with the reservation of a few places for 

 public monuments, is most convenient. This central part might be 

 surrounded by a boulevard, beyond which the circular type might be 

 adopted with modifications so as to avoid curved streets. The princi- 

 pal streets of each suburb might be directed toward the center of the 

 city, and each suburb might in itself be laid out more or less according 

 to the rectangular type. The transverse streets would, however, 

 spread farther apart as the distance from the center became greater ; 

 and main diagonal streets might be arranged to cross the whole city, 

 with few or only slight deviations. Beyond the suburbs, the principal 

 radial streets might be continued for a considerable distance farther, 

 but the transverse streets would nearly disappear. The accompanying 

 design has been drawn according to these principles. It is worthy of 

 remark that the sketch is more like the European cities that have 

 grown up by progressive additions than the American cities which 

 have been built on a so-called rational plan. 



So far we have considered the question from an exclusively geomet- 

 rical point of view. We may continue our study by referring to the 

 influences which geographical, meteorological, commercial, and polit- 

 ical conditions have had in the laying out of cities. Certain directions 

 in the ways of communication are often imposed by the topographical 

 situation ; as, for example, when the town is crossed by a river or a 



