98 TRANSACTIONS OF THE AMERICAN SOCIETY 



is 19/4 cents; in Boston 25 cents, or 1.28 times the Hartford rate. In making comparison 

 of the costs of maintenance per acre for a certain character of landscape treatment in the 

 parks of those two cities, with the purpose of getting some gauge upon the relative efficiency 

 of administration, the actual cost in Hartford should therefore be increased by 28 per 

 cent. If, in any case, after such an allowance is made for the difference in the price-factor, 

 there appears a considerable difference in costs, while it will not afford a positive and 

 direct measure of the relative efficiency of the two administrations, it wilt be a signal to 

 examine more critically the other factors; arid if the quality of upkeep appears from 

 personal observation to be no lower in the case where the cost is lower it indicates greater 

 net efficiency in one case than the other, and suggests an inquiry into the means by which 

 this greater efficiency is secured, means which are probably within the reach of the other 

 manager when his attention is thus called to them. 



It is this aspect of cost-accounts as furnishing an approximate gauge of the relative 

 efficiency of different methods employed in different places, or at different times in the 

 same place, or at least as calling attention to the particular points where comparison of 

 methods and results is most likely to be instructive, that chiefly interests superintendents, 

 park commissioners and owners of property, when they no longer have extensive improve- 

 ments to make, but are confronted with the problem of economical and efficient maintenance. 



To the landscape architect and to his client, when confronted with problems of design 

 and improvement, the interest lies upon the other side. The question is this: Assuming 

 normal conditions as to all three of the administrative factors (prices, quality of upkeep 

 and efficiency of management), what will be the effect on the cost of maintenance of 

 differences in such of the initial factors as are under the designer's control? Practically 

 the question presents itself in connection with two kinds of choice, the choice between 

 alternative methods of treating a given area and the choice as to the relative extent of 

 areas to be devoted to different kinds of treatment. For example, in a certain spot a "wild 

 garden" of carefufly chosen and skilfully arranged herbaceous plants would look very 

 well and be very enjoyable, but the same place would still be beautiful even though some- 

 what less attractive and interesting if planted with a simple mass of shrubs around a 

 piece of greensward. To decide wisely between these alternative methods of treatment 

 some one ought to consider what the difference in cost of maintenance will be, and whether 

 in that particular locality the more elaborate treatment would be enough pleasanter to 

 be worth the difference. On the other hand, it may appear that a design involving three 

 acres of formal garden, eight acres of lawn, and four acres of shrub- and tree-plantations, 

 is more attractive than another involving two acres of formal garden, five acres of lawn, 

 and eight acres of shrub- and tree-plantation, and the question is: About how much 

 more will it take to keep the former in good condition, and is it enough more attractive 

 to be worth the extra cost? 



In getting data for answering such questions, allowance can be made, as before, 

 for differences in the price-factor; but cases of abnormal quality of upkeep, either high 

 or low, and cases of either extreme inefficiency or very unusual efficiency must simply 

 be eliminated in getting at averages of cost. What is most instructive, however, is not 

 a meclianical average but a condensed statement of a number of cases with an indication 

 of the relative qualities of upkeep and degrees of efficiency as deduced from general 

 observation^ 



Of the initial factors, differences in the extent of the area to be maintained in any 



