Except in the Southwest operating costs per thousand feet were less in the country trade, while in all instances higher net profits are shown by country yards. The returns expressed in board-foot units, however, may not be a true index to the profit as a percentage on the investment in the business. By reference to Tables 11 and 14 it will be noted that an average gross profit of $7.25 per thousand feet in the Northwest yielded a return on the investinent of 12.5 per cent, while a gross profit of $7.07 per thousand feet in the Southwest yielded a net return on the investment of only 7.4 per cent. In Kansas City a gross profit of $5.65 per thousand feet yielded a return on the investment of 6 per cent; while in Minneapolis a gross profit of $5.85 per thousand feet was required to earn the same return. Considering the three large cities together in comparison with the outlying country districts, the figures indicate that the same gross margin per thousand feet would have earned approximately the same return on the investment. FLUCTUATIONS IN RETAIL PRICES AT LARGE CITIES. The average price received for all the lumber handled monthly or annually by a retail yard represents an average of many different grades and patterns sold at different prices and in varying proportions. Fluctuations in the average price may be due simply to the sale of varying proportions of different priced stock. The average price received, therefore, can not be taken as a measure of price fluctuations on the same grades of lumber or as indicative of the extent of competition. Lumber is priced at the mills according to its quality or grade, finish, and size. Differences in the length or width of pieces commonly carry differences in the price per thousand feet. Many of these fine distinctions disappear in the retail trade and more uniform prices are asked for lumber of standard lengths. Variations in retail prices on the same grade thus indicate either changes in the wholesale market or competition, or both. Manufacturers generally are of the belief that the retail prices of lumber seldom change and that the retailer reaps the benefits of low wholesale prices without passing them on to the consumer. In order to determine the extent to which retail prices are subject to fluctuations and change in sympathy with wholesale prices a number of cities and towns representing different sets of conditions were selected and the trend in retail prices during the past three and a half or four years ascertained for selected grades. Chicago, Kansas City, and Minneapolis were selected as representing the metropolitan market where many yards are located and keen competition would be expected. Prices covering a similar period were obtained at nine additional points, representing trade conditions in towns ranging in population from 6,500 down to 40. Of these points, five contained two or more yards each; the remaining four are one-yard points of less than 350 population. TREND OF PRICES AT CHICAGO. In figures 8 to 11 are given the price curves platted from data obtained at a number of yards in Chicago. By reference to figures 8 and 9 the basis for the curves and the manner in which they were constructed will be clear. They are drawn to represent an average of all the selling prices on the grades in question which were obtained from the different companies. The data were secured by going through the sales tickets of the several companies and recording the actual prices at which the grade of lumber was sold. Sufficient time was not available to record all the sales made or to weigh the prices according to their volume. A sufficient number of sales distributed throughout the month were selected, however, to reflect fairly the market prices at which the lumber was being sold. The sales were then averaged by months and the average price obtained by each company platted. Differences in the prices obtained by various com FIGS. 8 AND 84.-Course of retail prices, Chicago, as reflected by current sales. In fig. 8 the dots represent the monthly average of sales made by the different companies. panies held good on all the items selected, and uniform retail prices were not evidenced by any of the data. The retail dealers in Chicago issue a retail price list from time to time which they endeavor to follow, but the curves reflect very strikingly the extent to which they depart from it. Special effort was made to show the actual market prices as closely as possible. Discounts were deducted. The curves do not, however, represent the average prices prevailing in the market, since they are based on a limited number of sales and not on the total footage handled. The prices were obtained from four representative yards located in widely separated parts of the city. The amount of western lumber sold through retail yards in Chicago is relatively small and did not afford a sufficient volume upon which to base monthly and yearly price curves. Figures 10 and 11 reflect very strikingly the close relationship between wholesale and retail prices in a large city market like Chicago. TREND OF PRICES AT KANSAS CITY. Retail-price curves similar to those shown for Chicago are given. for Kansas City in figures 12 to 16, inclusive. These curves were constructed in the same manner as those obtained in Chicago (see figs. 12a and 13a); and it will be noted that they follow the same course and reflect practically the same competitive conditions. In both instances retail prices rose steadily during 1912 and reached their highest point about the middle of 1913, since when they show a consistent decline to 1915. Thus they conform closely to the movement of wholesale prices over the same period, which is shown graphically in figures 14, 15, and 16. A possible exception is No. 2 FIG. 9.-Course of retail prices, Chicago, as reflected by sales of three companies. The dots represent the monthly average of sales made by the different companies. Vertical Grain Flooring of Douglas fir, in which the fluctuation is not so extreme. This curve, however, is misleading, in that extreme fluctuations in the prices received by different companies are counterbalanced in the averages represented by the curve. Curves for. other grades of western species were not obtainable in Kansas City. TREND OF RETAIL PRICES IN MINNEAPOLIS. Retail prices at Minneapolis were obtained from two companies operating in different parts of the city and curves constructed in the same manner as for Chicago and Kansas City. Minneapolis being a northern pine market, it was necessary to select several grades of this species in order to reflect with approximate accuracy the fluctuations in monthly and yearly prices. Douglas fir flooring was the only grade of western lumber for which sufficient data could be obtained to justify a curve. The curves are given in figures 17, 18, and 19 following. It will be noted that the trend of northern pine retail prices in Minneapolis does not conform with that shown for Douglas fir or |