UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY E. RAUCHENSTEIN BULLETIN 419 February, 1927 UNIVERSITY OF CALIFORNIA PRINTING OFFICE BERKELEY, CALIFORNIA 1927 ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY E. KAUCHENSTEINi SUMMARY The cantaloupe industry has been expanding rapidly in the irri- gated areas of southwestern United States, especially in the Imperial Valley of California, The eastern and southern states are increasing only slightly in cantaloupe shipments. The shipments from Arkansas and Indiana, although comprising but a small proportion of the total United States shipments, have shown decided increases. Imperial Valley cantaloupes have ripened much earlier in 1924, 1925, and 1926 than in previous years. The increase in early ripen- ing has been attributed largely to the introduction of certain new varieties and to the use of brush shelter. However, there are two other factors which need to be considered in comparing earliness of ripening when judged, as it usually is, by the date when the first carlot was shipped out of the Valley. These factors are, first, the increase in acreage which makes it possible to get a carlot together earlier now than formerly, and second, the relatively high temperatures which have prevailed during the growing period in recent years compared with the earlier years. The latter is undoubtedly purely a matter of chance and cannot be expected to continue. The first cantaloupe shipments from Imperial Valley usually come onto the markets in June and miss the keenest competition from such other fresh fruits as peaches and watermelons. Thus prices are obtained for the early cantaloupes which have greatly stimulated their production in Imperial Valley. The number of carlots of cantaloupes shipped from Imperial Valley increased from 4446 in 1914 to 13,410 in 1925, and the total number of cities receiving over one carlot increased from 176 in 1914 to 373 in 1925. The average price of Imperial Valley cantaloupes during any season is closely related to the acreage planted in Imperial Valley the following year. This is especially marked for the years 1920 to 1926, for which it was possible to calculate weighted prices. The degree of relationship between weighted New York prices and the acreage of Associate in Agricultural Economics. 4 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION the following year is shown by the correlation coefficient of 0.98. (A correlation coefficient of 1.0 indicates perfect relationship.) In other words, years of high prices are followed by marked increases in acreage and years of low prices are followed by marked decreases in acreage. Approximately 20 per cent of Imperial Valley shipments are sold f .o.b. loading point, The f .o.b. prices for each individual year except for 1926 are closely related to shipments. However, there are rather wide variations between years, which make it inadvisable to work out a general estimating equation for estimating future prices from ship- ments. The records of individual years, however, should be of value in judging the course of future prices whenever conditions at that future time are similar to those of some past year. New York prices of cantaloupes are closely related to receipts, temperature, and time of the season. If we take (1) weekly averages of carlot receipts, (2) maximum temperatures with a three-day lag, and (3) the time of the season, and correlate these with average weekly prices for the years 1921 to 1926, we find that approximately 88 per cent of the variations in the average weekly prices are accounted for by these three factors. Estimated average weekly prices for the period 1921 to 1926 based on the average relationships between prices, receipts, temperatures, and time of season come fairly close to the actual prices. Actual prices paid tend to alternate above and below the estimated prices, which suggests that one reason for the differences between actual and estimated prices is that dealers in estimating the demand of the consumer tend to go from one extreme to the other. Assuming that the same relationship between prices, receipts, tem- peratures, and time will prevail in 1927 as prevailed during the years 1921 to 1926, we may use the average results of this period as a basis for estimating prices in 1927. This should be supplemented, however, by any other factors which come to the dealers' attention which are not included in this study but which may affect demand. Thus the quality of cantaloupes affects the demand, but no measure of quality has been available to be included in this study in estimating future prices on the basis of past records. Special weather conditions other than maximum temperature, prices and quality of competing fruits, and the buying power of the consumers, may also affect the demand of consumers. Bul. 419 ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY OBJECT OF THIS STUDY The object of this paper is to present and to analyze the available data that have an economic bearing on the cantaloupe industry in the United States in general, with special reference to the cantaloupe industry in the Imperial Valley. Those factors which affect prices of Imperial Valley cantaloupes and the effect of prices on subsequent acreage are analyzed in considerable detail with the hope of bringing about a more stable industry. Average Coriot Shipments of Ceni-aloupes for /9Z4 and /9Z5 by Points o-f Origin Each Do-h Represents lOO Car/ots deg/nn/ng ancf Endr'ng Do+e o-f Shipping Season Given Gy Sections Fig. 1. — Data from table 1. Most of the cantaloupes are now produced in the irrigated sections of the West, where the yields and color of the fruit are more uniform than in the East on non-irrigated land. The earliest cantaloupes are produced in the Imperial Valley of southern California. AREAS PRODUCING CANTALOUPES IN THE UNITED STATES Based on average carlot shipments for 1924 and 1925, the most important cantaloupe shipping areas in the United States are in the West, mainly on irrigated land. The Imperial Valley of California alone shipped 14,583 carlots, almost one-half of the total carlot ship- ments in the United States during these two years, and the state as a whole shipped 17,395 carlots, which was 58.9 per cent of the United States total. Colorado was second in shipments with 2938 carlots (see table 1 and fig. 1) and Arizona was third with 2461 carlots. Arkansas, b UNIVERSITY OF CALIFORNIA EXPERIMENT STATION Indiana, Maryland, Delaware, New Mexico, and North Carolina fol- lowed next in order, each with more than 500 carlots shipped annually. The acreage figures for 1919, shown in table 2, emphasize eastern production to a greater extent than is shown by the reports of carlot shipments. For example, Colorado and Arizona are second and third in shipments, but sixth and seventh, respectively, in acreage. Mary- TABLE 1 Average Carlot Shipments of Cantaloupes for 1924 and 1925 by States or Sections of Origin in the United States State or section California — Imperial Valley. California — Central District Colorado Arizona Arkansas Indiana Maryland Delaware New Mexico North Carolina Texas Georgia Nevada Washington Michigan Tennessee New Jersey Total Ship- ments, carlots 14,583 2,776 2,938 2,461 1,128 955 908 584 546 528 443 351 289 259 130 120 109 29,108 State or section Illinois Iowa South Carolina Oklahoma California — Northern District Virginia Missouri Kansas Kentucky Louisiana Florida New York Idaho Alabama Utah Oregon California — Southern District. Grand total Ship- ments, carlots 86 81 72 50 35 27 21 18 9 6 6 3 2 2 2 1 1 422 29,530 Compiled from Summary of western cantaloupe deals— 1925. p. 18. (1) Hereafter, superscript figures in parentheses will be used to refer to literature cited on last page of this bulletin.) land is third in acreage, but sixth in shipments. Indiana is fourth in acreage, but fifth in shipments. The differences in the order of importance as shown by shipments reported in the Bureau of Agri- cultural Economics, United States Department of Agriculture Canta- loupe Deals (table 1) and the 1920 census (table 2) are probably due to two factors. First, the yields through the Middle West and the East are somewhat lower than in the important irrigated sections. Second, in states that are close to large markets a considerable pro- portion of the cantaloupe crop may be hauled by trucks, and hence will not be reported in carlot shipments. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY TREND OF CANTALOUPE PRODUCTION IN VARIOUS SECTIONS OF THE UNITED STATES A general idea of the trend in cantaloupe production by states since 1899 may be obtained from the United States Census which is summarized in table 2. The acreages given for 1909 are not strictly comparable with those given for 1899 and 1919 since the latter two TABLE 2 Cantaloupe Acreages in - the United States by States Having More than 1000 Acres in 1919 State California Colorado Arizona Arkansas Indiana Maryland Delaware New Mexico North Carolina Texas Georgia Michigan New Jersey Illinois Virginia Florida New York Ohio '. All others Total United States. 1899 60,854 1909 764 5,784 21,470 2,329 3,970 4,007 117 452 3,300 2,388 994 8,999 3,517 3,855 4,182 3,461 3,516 4,665 358 1,163 2,500 638 491 1,022 1,729 2,256 2,130 5,781 1,685 2,093 1,707 3,022 1,659 2,231 1,829 2,347 6,548 3,861 4,231 3,646 2,332 1,481 3,028 969 1,383 2,087 3,408 1,079 1,533 978 1,091 2,256 1,665 1,446 16,736 10,189 9,351 52,419 1919 78,436 Acreages for 1899 from Twelfth Census of the United States 6:338; for 1909 from Thirteenth Census of the United States 5:668; and for 1919 from Fourteenth Census of the United States 5:819, 824. are based on acreages of cantaloupes or muskmelons raised for sale, while in 1909 the acreage was based on all farm reports showing more than one acre of truck crops. California shows the greatest increase in acreage of cantaloupes, viz., from 764 acres in 1899 to 21,470 in 1919. Colorado, Arizona, and Arkansas also show substantial increases during that time. Indiana, Maryland, and North Carolina show moderate increases, while Illinois, Virginia, Florida, and Ohio show decided decreases. 8 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION A rough measure of the trend of the cantaloupe industry in the United States during the last ten years can be obtained from the records of carlot shipments. Table 3 and figure 2 show the carlot ship- ments by states and by sections from 1916 to 1925. These data may be affected to some extent by the completeness of the returns on ship- ments. However, the comparison of trends between sections should be reasonably accurate. TABLE 3 Carlot Shipments of Cantaloupes by Sections or States of Origin, 1916-1925 State and section 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 Section 1: Imperial Valley 4,617 5,007 4,398 7,831 8,903 10,686 12,159 12,583 15,756 13,410 Section 2: Southwestern States — Arizona 839 3,472 1,740 316 51 1,215 3,251 1,898 139 227 1,169 2,450 1,818 36 256 9 1,832 4,179 3,132 36 378 123 1,164 4,197 2,454 48 937 22 1,474 2,491 3,215 74 421 162 1,558 3,145 4,420 164 275 186 1,208 2,866 2,195 107 364 387 1,864 2,536 2,654 231 518 416 3,057 California (excluding Imperial Valley) 3,086 3,222 Nevada 346 Mew Mexico 574 470 6,418 6,730 5,738 9,680 8,822 7,837 9,748 7,127 8,219 10,755 Section 3: East and South — 905 975 1,087 844 74 348 702 789 855 1,106 157 249 429 551 490 418 31 129 590 314 835 523 100 183 581 389 771 359 110 208 943 640 1,206 821 299 469 843 1,632 1,233 700 270 447 818 216 1,270 620 70 341 511 586 699 401 115 232 657 117 Maryland 1,116 655 South Carolina 29 509 4,233 3,858 2,048 2,545 2,418 4,378 5,125 3,335 2,544 3,083 Section 4: Middle West, Arkansas and Washington — 603 45 794 31 13 260 797 119 664 68 42 145 699 103 443 43 37 110 1,106 85 462 26 204 100 936 85 635 40 209 329 1,501 97 644 41 176 209 990 116 894 49 465 371 337 86 681 109 306 207 1,052 45 822 79 114 298 1,203 126 1,087 82 146 220 1,746 1,835 1,435 1,983 2,234 2,668 2,885 1,726 2,410 2,864 17,014 17,430 13,619 22,039 22,377 25,569 29,917 24,771 28,929 30,112 Years 1916-1922, except for Imperial Valley alone, from Colorado cantaloupe deal, 1923. p. 9. (2) Year 1923 from U. S. D. A. Yearbook 1924. p. 694. Years 1924 and 1925 from Summary of western cantaloupe deals— 1925. p. 18. (1) Imperial Valley shipments 1916-1923 from the Packer, May 15, 1926; years 1924, 1925 from Summary of western cantaloupe deals. (1) For 1916-1923 Imperial Valley shipments were subtracted from total for the state to obtain shipments for California excluding Imperial Valley. BuL.. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 9 Figure 2 shows that since 1920 the Imperial Valley has shipped more carlots of cantaloupes than any of the other sections. From a total of 4617 carlots in 1916, shipments have increased rapidly to a total of 15,756 in 1924, and 13,410 in 1925. In other words, the ship- ments of cantaloupes in the Imperial Valley during the past ten years increased approximately 200 per cent, while the population in the United States increased 13.1 per cent. Trend of Carlot Shipments of Cantaloupes prom Various Sections op the United States, 1916-1925 "Thousands Fig. 2. — Illustrating table 3. The Imperial Valley is the most important cantaloupe producing area in the United States and has increased the fastest. The early western irrigated sections have generally shown the greatest increase. In the Middle West, Indiana and Arkansas have shown the most important increases. Next in importance to the Imperial Valley is section 2, which includes the rest of California and all of the other southwestern states. Practically all of the cantaloupes in this area are grown under irriga- tion, and hence yields tend to be more uniform from year to year than in those sections that depend entirely on rainfall for water. The cantaloupes here are also generally free from the white color which develops on the underside of cantaloupes that lie on the wet ground. This advantage, in addition to the better grading and packing methods which are generally followed in the cantaloupe growing sections of the 10 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION southwest, probably accounts for the premium which these cantaloupes usually bring on the markets. The increase in shipments in this section from 1916 to 1925 has been more than 50 per cent. The Turlock district of California, the Rocky Ford district of Colorado, and certain parts of Arizona have been the most important shipping areas of this section. Section 3, which includes the important eastern and southern cantaloupe-growing states, shows no tendency towards an increase in shipments, although there may have been an increase in hauling by truck, especially in such states as Delaware and Maryland, which are near the large markets. The shipments in this section vary more from year to year than the shipments from the irrigated sections. Section 4, a somewhat scattered group, includes the most important middle-western states and Washington. Arkansas and Indiana furnish from 50 to 75 per cent of the carlot shipments of this section. This group, as a whole, shows a slight upward trend but supplies less than 10 per cent of the total United States carlot shipments. DISTRIBUTION OF CARLOT SHIPMENTS OF CANTALOUPES BY MONTHS The earliest shipments of cantaloupes of any importance in the United States come from the Imperial Valley. In this district, which is favored by high temperatures, special attention has also been given to obtaining early varieties and to developing cultural practices which will bring cantaloupes to market at an early date. A comparison as to time of shipments of the various sections in 1925 is shown in table 4 and figure 3. Of the 1503 carlots of cantaloupes produced and shipped in the United States in May, 1925, all but 26 came from Imperial Valley. Texas produced the other 26 carlots. During June came the peak of shipments for the United States with a total of 10,078 carlots, of which Imperial Valley shipped 9745. In July, section 2, the rest of the southwestern area, took the lead in shipments with a total of 4782 carlots, which was its peak of shipments. Imperial Valley shipped 2187 carlots; section 4, 1667 carlots; and section 3, 1079. With shipments from Imperial Valley practically over by August, the total United States shipments declined to 6541 carlots. Section 2 supplied 3603 carlots, and sections 3 and 4 supplied 1871 and 1066 respectively. The September shipments amounted to only 2126 carlots, of which Colorado (section 2) supplied 1757, and Nevada (section 2) supplied 159. In October, Colorado shipped 141 carlots and Wash- ington (section 4) shipped 4. Caelot Shipments of Cantaloupes by Sections of Origin, 1925 Thousand CoHo-f-s Section I Imperial l/oiley, California > \ / / / V \ / \ ...---' .-'" \ / — N • r i >> * » \ i Section Z Stares Included- 4 Arizona Calif, (omitting L Vol.) Colorado Nevada 2 New Mexico Texas Section 3 States inchded- De/avsare Maryland North Carolina South Carolina Georgia Other States Section 4 States included - & Arkansas Illinois Indiana lo\*sa _- Michigan Washing-tan Fig. 3. — Illustrating sectional totals of table 4. The earliest important ship- ments of cantaloupes come from Imperial Valley in May. In June they exceed the total shipments from all other areas for any month of the season. In July, August, and September, section 2, the southwestern states, furnishes the bulk of the shipments. 12 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION TABLE 4 Monthly Carlot Shipments of Cantaloupes by Sections or State; of Origin, 1925 April May June July Aug. Sept. Oct. Section 1: Imperial Valley, Calif 1,477 9,745 2,187 1 Section 2: Arizona 214 2,824 1,568 19 1,472 1,324 115 372 301 California (excluding Imperial Valley) 46 1,757 159 Colorado 141 Nevada 72 202 116 New Mexico Texas 4 26 16 7 Total 4 26 230 4,782 3,603 1,969 141 Section 3: Delaware 32 40 151 603 26 227 618 1 942 52 7 Georgia 76 Maryland 23 North Carolina South Carolina 3 14 Other states 258 10 Total 93 1,079 1,871 40 Section 4: Arkansas 10 1,083 66 518 110 60 569 69 98 160 Illinois Indiana Iowa 13 48 56 Michigan Washington 4 Total 10 1,667 1,066 117 4 Grand total 4 1,503 10,078 9,715 6,541 2,126 145 Data from Summary of western cantaloupe deals, 1925. p. 18. (1) BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 13 FACTORS AFFECTING THE TREND IN EARLY SHIPMENTS OF CANTALOUPES IN IMPERIAL VALLEY During" the last few years there has been a striking increase in the earliness of ripening of cantaloupes in Imperial Valley, especially if judged by the time when the first carlot was shipped (see table 6). Thus in 1922 the first carlot was shipped May 31 ; in 1923, May 23 ; in 1924, May 19; in 1925, May 7; and in 1926, May 1. The increase in early ripening of cantaloupes in 1925 and in 1926 has been attributed largely to the introduction of certain new varieties such as "HB," and to the use of brush shelter. Undoubtedly these factors have influenced the earliness of ripening, but at least two other factors have helped to make it possible to ship the first carlot at an earlier date during the past few years than during the earlier history of the industry. These are, first, the increase in acreage, and second, the high temperatures which have prevailed during the growing period in recent years compared with the earlier years. Increase in Acreage. — In 1913 there were 6500 acres of cantaloupes in the Imperial Valley compared with 35,301 acres in 1926 (see table 10). Obviously it is possible to get a carlot together somewhat earlier from a large acreage than from a small acreage. This factor can be corrected by comparing shipments from year to year on the basis of carlots to 1000 acres or some other common base. In this study, comparisons are made on the basis of 1 and 5 carlots to 1000 acres. Increase in Temperature during the Growing Season of the Later Years Compared with the Earlier Years. — Temperature records from a government weather station in Imperial Valley are available since 1913. The early cantaloupes are usually planted in December and early January, and are covered with oiled paper until some time in March, when danger from frost is practically past. Growers estimate that in 1913 about one-half of the total cantaloupe acreage was covered in this way. In 1926 nearly two-thirds of the total acreage was covered. Under these conditions the temperatures from January to May of each year undoubtedly affect the time of ripening. Table 5 gives the mean monthly temperatures from January to May for the years 1913 to 1926. Fifty degrees Fahrenheit is approximately the lowest temperature at which cantaloupes make any growth. By adding up the mean temperatures above 50 degrees for January, February, March, and April, and multiplying this sum by 30 (the average number of days 14 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION for the first four months) we can obtain a close approximation to the accumulated daily temperatures above 50 degrees from January 1 to April 30. Multiplying the mean May temperature above 50 degrees by 30 gives a close approximation to the accumulated temperature to May 30. This gives a good basis for comparing the temperatures and earliness of ripening for the years 1913 to 1926. Table 6 shows the accumulated temperatures and various rates of shipments. TABLE 5 Mean Monthly Temperatures in Degrees Fahrenheit, Imperial Valley, 1913-1926 Jan. Feb. March April May 1913 49.5 56.4 59.7 69.0 75.6 1914 54.7 59.2 64.9 65.7 75.4 1915 53.4 61.8 63.4 69.0 73.3 1916 52.2 57.4 67.8 71.8 75.8 1917 50.4 57.4 58.8 65.8 71.0 1918 53.7 57.1 64.7 69.6 73.0 1919 51.6 56.0 60.4 71.8 77.0 1920 55.6 60.4 61.0 68.0 76.4 1921 52.2 59.2 66.2 67.0 72.5 1922 49.3 55.6 58.8 65.1 77.1 1923 56.1 56.8 60.4 67.8 78.1 1924 53.6 64.2 66.0 70.4 79.8 1925 53.2 61.2 64.4 70.8- 78.6 1926 53.9 61.5 66.9 73.0 77.5 Data from U. S. Weather Bureau Climate-logical Data, California Section 1913-1926. Temperature records for 1913 and 1914 for Brawley, 1915-1925 for Calexico, and for 1926 for Imperial. A close study of table 6 shows a fairly close relation 2 between accumulated daily temperatures from January 1 and dates of ship- ments. That is, in general, low temperatures go with late shipments, and high temperatures go with early shipments. Comparing columns II and V it will be noted that in 1913 the low accumulated tempera- tures of 1038 degrees Fahrenheit are associated with June 8, the date when cantaloupe shipments in Imperial Valley reached the rate of one car to 1000 acres. In 1917 the temperatures of 972 degrees are associated with June 11, and in 1922 temperatures of 864 degrees and the late date June 7 are associated. On the other hand, in 1916 the high accumulated temperatures of 1608 degrees are associated with an early date, May 22. The temperature of 1626 degrees in 1924 are 2 The correlation coefficient between accumulated temperature to April 30 and number of days after that until shipments reached 5 carlots to 1000 acres is — 0.76. Bul. 419 ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 15 associated with May 24; in 1925, 1488 degrees are associated with May 22 ; and in 1926, 1659 degrees goes with May 7. Summing up the accumulated temperatures above 50 degrees Fahrenheit from January 1 to the date when shipments reached the TABLE 6 Accumulated Daily Temperatures above 50 Degrees Fahrenheit after January 1 of Each Year to Dates Indicated, and Date; of Various Bates of Shipment Accumulated daily temperatures above 50 Date of shipments degrees Fahrenheit First carlot One car to 1,000 acres Five cars To Apr. 30 To May 30 to 1,000 acres I II III IV V VI 1913 1,038 1,806 June 4 June 8 June 11 1914 1,251 2,013 May 16 May 24 May 28 1915 1,350 2,049 May 25 June 1 June 3 1916 1,608 2,382 May 16 May 22 June 4 1917 972 1,602 June 4 June 11 June 15 1918 1,353 2,043 May 25 May 30 June 4 1919 1,194 2,004 May 26 May 30 June 2 Mean (1913-19) 1,252 1,986 1920 1,350 2,142 May 21 May 27 May 30 1921 1,338 2,013 May 21 May 27 May 29 1922 864 1,677 May 31 June 7 June 10 1923 1,281 2,076 May 23 May 27 May 29 1924 1,626 2,520 May 19 May 24 May 28 1925 1,488 2,346 May 7 May 22 May 28 1926 1,659 2,484 May 1 May 7 May 14 Mean (1920-26) 1,372 2,180 Temperature data compiled from U. S. Weather Bureau Climatological Data California Section, 1913-1926. Data on shipments 1913-1920 compiled from The Packer, May 28, 1921; for 1921-1925 from Summary of western cantaloupe deals— 1925. pp. 29, 30. (1) For 1926 from Preliminary Review of the 1926 Imperial Valley Cantaloupe Deal. p. 4. (3) rate of one carlot to 1000 acres shows, in another way, the relation between temperature and earliness of shipments. This is shown in table 7. The extreme range in temperatures accumulated to the date of shipment is from 1809 to 2320 degrees Fahrhenheit, which occurred in the years 1926 and 1924, respectively. The year 1926 is marked by the fact that the rate of shipments of one carlot to 1000 acres was 16 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION reached with the lowest accumulated temperatures of any year of the fourteen. However, 1914 was a close second with 1820 degrees, and 1922 third with 1871 degrees. During these years other factors than mean temperature which affect the time of ripening must have been exceptionally favorable to early ripening. In 1926 the new early variety undoubtedly contributed something to early ripening. The TABLE 7 Kelation Between" Accumulated Temperatures above 50 Degrees Fahrenheit and earliness of shipments Accumulated temperatures above 50 degrees from Jan. 1 Date when shipments reach to date, not including 1 carlot per 1,000 acres day of shipment 1913 June 8 2,017 1914 May 24... 1,820 1915 June 1 2,082 1916 May 22 2,173 1917 June 11 1,929 1918 May30! 2,022 1919 May 30 1,985 1920 May 27 2,012 1921 May 27 1,906 1922 June 7 1,871 1923 May 27 1,964 1924 May 24 2,320 1925 May 22 2,051 1926 May 7 1,809 Mean accumulated temperature 1,997 Mean 1913-1919 2,004 Mean 1920-1926 1,990 Compiled from U. S. Weather Bureau Climatological Data, California Section, 1913-1926 reports. practice of brushing a part of the fields, and the freedom from cold heavy winds in March are held, by some growers, to be contributing factors. The extent of the variations from the mean of 1997 degrees in table 7 indicates that mathematically the chances are even that the date when cantaloupe shipments reach the rate of one car to 1000 acres will come within the time when the accumulated temperatures above 50 degrees after January 1 will have reached the sum of from 1907 to 2087 degrees as long as cultural methods and earliness of the chief varieties remain about as they are at present. During the past fourteen years, eight years have come within this range. Bul. 419 ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 17 COMPETITION OF CANTALOUPES WITH OTHER FRESH FRUITS Cantaloupe shipments reach their peak in June (see fig. 4 and table 8). Practically all of the shipments during this month come from Imperial Valley. No other fresh fruit of any importance reaches its peak of shipments at this time. Fresh peaches do not compete seriously with cantaloupes until July, when the total shipments of Monthly Carlot Shipments of Specified Fresh Fruits in the United States, Average of 1918-1923, and 1924 and 1925 Car/a** Corrt-a/ocipee IOOOO Strawberries ZOOOO IOOOO Watermelons ZOOOO IOOOO Grapes Q IOOOO Oranges Apples 40000 3O0OO ZOOOO to OOO o / v a c 19 \ 8-1 9Z3 s / 1 J 1 w*BS£ — ■ / / 19 24 I \ / \ N ***, / f / / / ■— r-^ ««— i / \ \ \ I I f \ % I 19 Z5 \/ / / / -^- - — — f 4 r 1 1 K I \ 1 Fig. 4. — Illustrating table 8. Cantaloupe shipments reach their peak in June. Peaches and watermelons reach their peaks in July. Grapes and apples reach their peaks in the fall. Thus the June cantaloupes are relatively free from competition with other important fresh fruits. peaches usually exceed those of cantaloupes. By this time, however, most of the Imperial Valley crop has been marketed. Strawberry shipments reach their peak in May (see fig. 4 and table 8) although there must be large quantities of them hauled by truck in the east during June. The average carlot shipments during June from 1918 to 1923 numbered 2691, compared with 7767 carlots of cantaloupes. 18 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION TABLE 8 Carlot Shipments of Specified Fresh Fruits in the United States Average 1918-1923, 1924, and 1925 Fruits Cantaloupes.. Peaches Strawberries. Watermelons. Oranges Grapes Apples Cantaloupes.. Peaches Strawberries.. Watermelons. Oranges Grapes Apples Cantaloupes.. Peaches Strawberries.. Watermelons. Oranges Grapes Apples Average, 1918-1923 Apr. 1,504 5,962 May 308 603 7,159 1,133 4,870 June 7,767 3,117 2,691 8,669 3,705 12 258 July 6,877 8,387 180 9,489 2,684 417 1,910 Aug. 5,377 8,619 48 9,771 1,959 4,151 3,527 Sept. 2,535 8,195 21 1,893 1,775 17,180 12,506 Oct. 299 645 2 81 1,351 19,588 35,276 1924 1,425 9,788 1,281 28 10,093 65 6,682 2,295 11,863 1,873 6,188 6,602 5,789 7 912 8,311 14,599 764 26,024 3,162 1,251 2,362 4,592 13,683 10,470 2,917 7,447 3,126 2,697 7,889 2,458 2,604 26,215 14,641 187 1,322 120 2,353 28,892 39,866 1925 2,976 5,714 1,480 335 7,387 571 4,986 1,525 10,078 4,730 1,047 10,959 3,321 60 889 10,006 16,879 31 16,430 2,003 1,267 2,667 6,455 9,505 10,643 1,734 5,261 3,958 2,023 7,334 2,254 1,719 34,422 19,082 113 288 41 1,175 33,555 41,323 Data on cantaloupes, peaches, strawberries and apples for 1918-1923 compiled from U. S. Bur. of Agr. Econ. Statistical Bui. 7:6, 16, 25, 34; for 1924 and 1925, including also watermelons and oranges, from U. S. D. A. Crops and Markets Monthly Sup. 2:196, 197, 234, 235, 272, 306, 338, 370. Data on watermelons, 1921-1923, and for oranges and grapes for 1919-1923 compiled from U. S. D. A. Weather Crops and Markets 4:225, 305, 420, 602, 651. Data on watermelons 1919, 1920 from West, Carl J., and Lewis B. Flohr. Market statistics U. S. D. A. Bui. 982:242. Data on grapes compiled from U. S. Bur. of Agr. Econ. unpublished data. During the years 1918 to 1923 the carlot shipments of watermelons in May averaged 1133, compared with 308 carlots of cantaloupes for the same period. During 1924 and 1925 the cantaloupe shipments in May exceeded the watermelon shipments by approximately 1000 carlots, but these were unusually warm years for Imperial Valley, as is shown in table 6. During June the watermelon shipments usually exceed cantaloupe shipments and reach their peak in July or August. Peach shipments from 1918 to 1923 exceeded cantaloupe shipments BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 19 during May, July, August, and September. During June cantaloupe shipments were more than twice as high as the peach shipments. Grapes do not come on the market in large quantities until August, and hence compete only with late cantaloupes. Orange shipments have an earlier peak than cantaloupes but are available at practically all times and at all places. The competition with cantaloupes is, therefore, probably more serious than is indicated by the low shipments during the cantaloupe season. Apple shipments do not reach their peak until late in the fall, when the shipments exceed the combined shipments of all of the fruits shown in figure 4 and table 8 The number of carlots of apples shipped is not a good index of apple consumption because of the possibility of storing the crop. They fill a somewhat different place in the diet of most people, however, and hence are not serious competitors of cantaloupes. Other crops such as mixed melons, plums and prunes, cherries, and grapefruit, show relatively small shipments compared with cantaloupes, and hence can compete only to a limited extent with cantaloupes. TABLE 9 Number of Markets Reached and Carlot Shipments of Imperial Valley Cantaloupes, 1914-1925 1914 1915 1916 1917 1918 1919 Number of carlots shipped. ... Number of cities receiving over ten carlots 4,446* 691 176$ 4,729 56 157 4,617 5,007 4,398 7,831 74f 213f Total number of cities re- ceiving over one carlot 1920 1921 1922 1923 1924 1925 Number of carlots shipped. ... Number of cities receiving over ten carlots 9,015 84 231 10,709 1 94 262 12,243 113 331 12,997 112 315 15,930 136 390 14,509 118 Total number of cities re- ceiving over one carlot 373 • The Packer. May 28, 1921. t Marketing and distribution of western muskmelons in 1915. U. S. D. A. Bui. 401:21, 22. (4) t Summary of western cantaloupe deals— 1925, pp. 52, 53. (1 ^ 20 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION INCREASES IN NUMBER OF MARKETS FOR IMPERIAL VALLEY CANTALOUPES With the rapid increase in the carlot shipments of cantaloupes from Imperial Valley there has come a need for extending the markets. That the relative increase in the number of cities reached since 1919 very nearly kept pace with the increase in the number of carlots shipped during the same period can be seen by a glance at figure 5 and table 9. In 1919, 7831 carlots were shipped out of Imperial Valley, and in 1925, 14,509 carlots were shipped — an increase of 85 per cent. The total number of cities receiving more than one carlot of cantaloupes from Imperial Valley in 1919 was 213, and in 1925 it was 373, an increase of 75 per cent. Cities receiving more than ten carlots numbered 74 in 1919, and 118 in 1925, an increase of 60 per cent. This indicates a fair growth of sales in markets already estab- lished and a fair expansion into new markets. Carlot Shipments of Cantaloupes from Imperial Valley and Number of Cities Reached with Carlots No. of No. o-f Cor/ofis Cities 30 OOO 300 20 OOO ZOO Tb+aJ Num. bar o e at '*%- ^ mm ^^+ ^ 10 OOO too *£& .___- **? >+ 6 OOO 0 A OOO 40 Z OOO 20 / ooo to i £&* Mt i. or c 'frtes l On tO Car/on h 9 ~NumL lmp& ten an r~/o/ V c Car ?&por i-ed 3 5 & $ * < 1 < \ 5 | 5 I \ 1 \ ! 3 i i I I 1 Fig. 5. — Illustrating table 9. The number of carlots of cantaloupes shipped from Imperial Valley, 1914-1925, shows a fairly constant percentage increase from year to year. The number of cities reached increased only slightly less rapidly. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 21 EFFECT OF AVERAGE PRICE PAID FOR IMPERIAL VALLEY CANTA- LOUPES ONE YEAR ON THE ACREAGE OF THE NEXT YEAR Most products respond to price changes. That is, an increase in price leads to an increase in supply and a decrease in price leads to a decrease in supply. The time required for this response depends largely upon the nature of the product. Tree fruits, which do not come into bearing for five or ten years after planting, naturally respond more slowly to changes in price than do annual crops like cantaloupes. The response in the case of cantaloupes is best shown by the changes in acreage the year after a change in price. Relative Purchasing Power of Imperial, Valley Can and Acreage, 1911-1926 TALOUPES, 1910-1925, Acres Purch- Poorer 1 / ~t 1 pa no \ / -Pure hcrsi ig f 'owe - h v\ t \ \ 1 \ / \\ f 1 V \ // Z4 90 fl V 1/ 1 1 - 1 r 1 I 1 Acr* >oc? t N 1 1 \ 7 1 1 \ / * / o Yr-s. -Tar- Par. P. f9IO II Yr S . ibr Acrea 3 el9lt fZ £3 Z4- /9Z5 Z6 Z4- t9ZS Z6 Fig. 6. — Illustrating table 10. The relation between the purchasing power and the acreage of the following year is not so clearly shown during the early years when the cantaloupe acreage was rapidly expanding. Since 1920, how- ever, years of high purchasing power have been followed consistently by increased acreages, and years of low purchasing power by decreased acreages. 22 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION Figure 6 and table 10 show the relative purchasing power of Imperial Valley cantaloupes for "Standard 45 V from 1910 to 1926. The price data for 1910 to 1919 are taken from the weekly quotations in the Pacific Rural Press. Simple averages of the weekly quotations were taken to represent the average price for the season. The price data for 1920 to 1922 and for 1925 and 1926 were taken from the New York Journal of Commerce Daily. For 1923 and 1924 daily prices were obtained from the Market News Service, New York Office of the Bureau of Agricultural Economics, United States Department of Agriculture. The price at New York each day for the years 1920 to 1926 was multiplied by the carlot receipts of Imperial Valley ' TABLE 10 Prices and Purchasing Power of Imperial, Valley Cantaloupes and the Acreage of the Following Year, Year Price per standard crate Index number Purchasing power in cents Relative purchasing power Year Acreage in Imperial Valley I II III IV V VI VII 1910 $2.29 2.13 1.91 2.25 2.21 1.44 2.16 2.16 3.06 3.64 4.51 3.78 2.71 3.74 2.92 4.05 3.32 103 95 101 102 100 103 130 181 198 210 231 150 152 157 153 162 153 222 224 189 221 221 140 166 119 155 173 195 252 178 238 191 250 217 120 121 102 119 119 76 90 64 84 94 90 116 82 109 88 115 100 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 4,500 1911 5,500 1912 6,500 8,200 8,300 9,200 13,800 1913 1914 1915 1916 1917 10,900 1918 1919 14,000 21,370 21,350 1920 1921 31,050 1922 24,863 1923 31,075 1924 27,559 1925 35,301 1926 Column II. Years 1910-1919 simple averages of weekly quotations compiled from Pacific Rural Press, current issues. For 1920-1922 and 1925, 1926, weighted average prices based on quotations in New York Journal of Commerce, Daily, current issues; 1923 and 1924 weighted average prices based on mimeographed reports of U. S. Bur. of Agr. Econ. Ww V, \-^-\ V --W ^ %5h ^ s * V* f> . TimtWkt t of Season Vrars - Z 3 4. I9ZI Z 3 4 S S /&S2 ■3 t S & /9Z3 ! z 3 a- s « I9R4- / Z 3 A. S /9Z5~ i z 3 a s « /see Fig. 8. — Illustrating table 16. There is a tendency for actual prices, where they do not agree closely with estimated prices, to alternate above and below the estimated prices. This may, in part, be due to the general tendency on the part of dealers to go from one extreme to the other in judging the demand of the final consumer. no receipts, and was not included. The following six weeks in each season, 1921 to 1926, were included. The remainder of each season of Imperial Valley cantaloupes, if any, was omitted because of the irregular receipts and lack of uniformity in the quality of the canta- loupes that often come on the market toward the end of the season. Due to the fact that the value of the dollar changed to some extent during the period 1921 to 1926, it was necessary to correct for these changes in studying the relation between prices and the factors affect- ing prices for the whole period. 5 The estimated prices, however, in 5 Changes in the value of the dollar from year to year were adjusted by divid- ing all prices by the Bureau of Labor Statistics index number, 1910-1914 base, and then converting the estimated prices back to the original value of the dollar for each year, 1921-1926. 34 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION TABLE 16 Belation of New York Prices for Imperial, Valley Cantaloupes to New York Beceipts, Temperature, and Time Date, week ending Average number of carlots daily Temperature in degrees Fahrenheit (3-day lag) Time- week of the season Actual price Estimated price Index No. I II III IV V VI VII 1921— June 12 13 76 1 $6.44 $6.45 18 21 81 2 4.72 5.56 25 30 87 3 3.77 5.02 July 2 40 82 4 4.34 4.02 9 55 87 5 3.61 3.78 16 46 83 6 3.15 3.48 150 1922— June 10 2 81 1 10.30 12.52 17 6 77 2 9.09 • 7.72 24 27 75 3 4.44 4.32 July 1 37 79 4 3.40 3.97 8 66 78 5 3.23 3.13 15 66 82 6 2.04 3.12 152 1923— June 9 11 82 1 8.11 7.83 16 41 71 2 3.92 4.03 23 48 82 3 4.23 4.19 30 50 88 4 5.52 4.33 July 7 62 75 5 3.08 3.19 14 63 81 6 2.59 3.20 157 1924— May 31 3 65 1 8.20 8.58 June 7 15 71 2 5.56 5.34 14 45 68 3 3.67 3.32 21 56 74 4 2.61 3.24 28 72 81 5 2.69 3.23 July 5 68 75 6 3.18 2.77 153 1925— May 31 5 70 1 8.50 8.44 June 6 11 79 2 7.66 7.08 13 43 88 3 4.75 4.94 20 61 77 4 3.92 3.52 27 60 80 5 3.59 3.54 July 4 77 77 6 3.75 2.92 162 1926— May 15 7 62 1 6.50 6.32 22 8 70 2 6.04 6.35 29 36 65 3 3.79 3.38 June 5 64 67 4 2.30 2.78 12 60 65 5 2.90 2.63 19 66 76 6 3.29 2.85 153* * Estimated for June, 1926. Columns II and V. Years 1921, 1922, 1925, 1926, compiled from New York Journal of Commerce, Daily, current issues. Years 1923 and 1924 from Review of season for California cantaloupes on New York mar ket. (6) - (7) Column III. Years 1921-1925 compiled from U. S. Weather Bureau Climatological Data— New York Section. Year 1926 compiled from New York Journal of Commerce, Daily. Column VI. Estimated prices based on average relation between columns II, III, IV, and V. The method of multiple correlation is used which is described in Mills, F. C, Statistical Methods, pp. 485- 515;^ 5 - ) and Wallace, H. A., and Geo. W. Snedecor. Correlation and machine calculation, pp. 18-47.^^ Column VII. Bureau of Labor Statistics index number, converted to 1910-1914 base. Cornell University, Farm Economics. 34:441. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 35 column VI, table 16, and shown by the broken line in figure 8, have been converted back to dollars of the year in which they appear, so that they are comparable with the actual prices paid as shown in column V of table 16. In the case of New York prices several factors besides receipts have had measurable effects upon them. These factors are tempera- ture and the time of the season. That is, high temperatures increase demand, other factors remaining constant, but toward the end of the season the cumulative effect of cantaloupe receipts, or perhaps to some extent the increased quantities of other fresh fruits and vegetables which arrive as the season progresses, cause lower prices for canta- loupes than would be expected on the basis of receipts alone. The combined correlation, or multiple correlation index, as it is technically called, of receipts, temperatures, and time of season with price for the period 1921 to 1926 was 0.94 as compared with a possible multiple correlation index of 1.0. Table 16 shows the average daily receipts of cantaloupes at New York, maximum temperatures three days earlier, time, actual and estimated price per standard crate (salmon tint 45 's), and the index number. The maximum temperatures prevailing three days before a given day were more closely related to the price of that day than concurrent temperatures or temperatures one day earlier. In this study temperatures three days earlier were taken. On the basis of the average relationships between receipts, temperature, time, and price, the estimated prices were calculated, shown in table 16, column VI, and by the broken line in figure 8. A close study of figure 8 and columns V and VII of table 16 indicates that the actual prices for each season as a whole are very close to the estimated, but for individual weeks occasional deviations of actual from estimated prices occur, amounting to 10 and 20 per cent. In 1921, during the first week, the actual price was $6.44, com- pared with an estimated price of $6.45. In other words, on the basis of the average relations that prevailed during the cantaloupe seasons of 1921 to 1926, between receipts, temperatures, time of the season, and prices, a price of $6.45 would have been expected, and the actual price paid was only one cent below that, or $6.44. During the second and third weeks the actual prices went con- siderably below the estimated prices. The fourth week of the 1921 season was marked by actual prices which were above the estimated prices. During the fifth and sixth weeks, however, the actual prices were again slightly below the estimated. 36 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION The 1922 cantaloupe season in New York was marked by alter- nating periods in which actual prices were first below the estimated then above during the second week, practically agreeing the third week, below during the fourth and sixth weeks, but agreeing the fifth week. Prices of oranges were high during 1922, which might lead one to expect higher than average prices for early cantaloupes. Pos- sibly the fact that the Imperial Valley acreage of cantaloupes was almost 50 per cent greater than the previous year may have led dealers to pay lower than normal prices during the first week in continual anticipation of heavy shipments later. During the second week, with shipments still continuing light, the actual price was 18 per cent above the estimated as compared with 18 per cent below the estimated for the first week. During the next three weeks actual and estimated prices agreed fairly well. During the last week the actual price was more than one dollar below the estimated. This may have been due to the unusually heavy cantaloupe shipments from other parts of the United States. The first three weeks of 1923 receipts of cantaloupes at New York resulted in actual prices very close to the estimated prices. The fourth week, however, actual prices exceeded estimated prices by $1.19. Two weeks later actual prices averaged $0.61 below the estimated. No explanation can be found for the unusually high price during the fourth week. To be sure, the temperature, lagged three days, was unusually high (88 degrees Fahrenheit) and might have had more than a proportional effect on demand, but the third week in 1921 a temperature of 87 degrees Fahrenheit was accompanied by an actual price $1.25 below the estimated. The third week in 1925 a tempera- ture of 88 degrees was accompanied by an actual price $0.19 below the estimated. These results are not consistent enough to indicate any unusual effect of exceptionally high temperatures on prices. In 1924 the agreement between actual and estimated prices was close during the first three weeks. The fourth and fifth weeks resulted in low actual prices compared with estimated prices. This condition was reversed in the sixth week. During 1925 the actual prices in the second, fourth, and sixth weeks were above the estimated, and only during the third week was the actual price slightly below the estimated. A simple average for the six weeks shows that the actual price was $0.29 above the estimated price. Several factors may have contributed to the favorable prices. 1. The quality of cantaloupes was generally very good. 2. An advertising campaign had been carried on to reach the consumer. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 37 3. The production of early cantaloupes in the United States as a whole was only 5,304,000 crates in 1925 compared with 6,210,000 in 1924. 4. Prices of oranges were relatively high in 1925. However, they were still higher in 1922, a year in which the simple average of the actual price was $0.38 below the simple average of the estimated price. In 1926 the actual and estimated prices agreed fairly closely for the first three weeks. During the fourth week the actual price was below the estimated, which situation was reversed during the fifth and sixth weeks. The unprofitable deal of 1926 was not due to unusually low prices, but to the low yields. In general, the agreement between the estimated prices of Imperial Valley cantaloupes on the New York market based upon the average relationship of receipts, temperatures, and time of the season to price for the seasons 1921 to 1926 is fairly close. About 88 per cent of the variations in price are explained by these three factors. Some of the factors affecting prices of Imperial Valley cantaloupes are: 1. The prices and quality of competing fruits and vegetables. 2. Weather factors in addition to temperature such as humidity and sunlight. For example, a clear, bright day with low humidity would undoubtedly stimulate demand for cantaloupes more than a cloudy day with high humidity having the same maximum tempera- ture in the shade as taken by the weather stations. 3. The buying power of the consumers is important and may change at times so as to affect the demand for cantaloupes. 4. Dealers cannot estimate correctly at all times what the demand will be, hence actual prices will fluctuate above and below the price which will keep supply and demand in equilibrium. Some factors affecting cantaloupe prices cannot be expressed in figures, others affect the price only occasionally so that no reliable average of their effect can be obtained. Besides the factors noted above and other factors which may be present that also affect the demand for cantaloupes, there are defects in the data and in the methods of calculating the estimated prices which would prevent securing a perfect correlation of actual and estimated prices, even though all of the factors affecting demand were known and could be definitely measured. No market reporting service, however perfect, can express in one figure an absolutely representative price for all of the sales taking place in one day or week. All that can be hoped for is a quoted price which represents a close approximation to the representative price. With the accumulation of more accurate data 38 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION from year to year and improved methods of analysis, it will be possible in the future to make more accurate estimates of the prices that are justified with a given supply and the known factors affecting demand than have been possible so far. However, the successful dealer will not depend entirely upon the analysis of the important measurable factors that have affected prices in the past, but will always be on the lookout for new factors, though they may be of only a temporary nature. A close-up view of any business will reveal many factors which cannot, or at least, have not been reduced to figures, but which do affect prices. These need to be evaluated and require the exercise of the dealer's best judgment. No statistical analysis of past data can be depended on as an absolute guide for estimating future prices, but it is valuable if supplemented by the judgment of one who is in close touch with the business. HOW TO USE THE RESULTS OF THE PAST SIX YEARS IN ESTIMATING FUTURE PRICES It must be clearly understood at the beginning that if we use the average results of the three factors related to prices described in the previous section in estimating the price of cantaloupes on the New York market for any time in the future, say 1927, we are assuming that these factors, receipts, temperatures, and time will continue to affect prices as they have done during the past six years. There is a fair degree of probability that the relationships will be the same as they have been, but there is no certainty that this will be the case. Having the assumption in mind, we may proceed to the task of esti- mating cantaloupe prices in New York for the season 1927. In order to abbreviate the description, it is advantageous to designate each factor used by a letter. The following symbols are used throughout this discussion : X = estimated purchasing power 1910-1914 base A = average daily receipts in carlots b = deviations from the average 1921-1926 maximum tempera- ture for the week, lagged three days C = time in weeks of the season. In order to convert purchasing power, X, into 1927 dollars it will be necessary to estimate the Bureau of Labor Statistics index number for 1927. This can be done more accurately before the cantaloupe season opens than at the time this publication is being written. How- ever, since the price level has been fairly stable since 1921, we are not likely to err much in estimating an index number of 155 for 1927. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 39 Estimating Prices by Means of the Logarithmic Regression Equa- tion. — Those readers who have some knowledge of logarithms as given in algebra texts will understand the statement of the equation for estimating purchasing power of cantaloupes in the following form: log ~X = 0.9686 — 0.3036 log A + 0.0071& — 0.0210C. Those readers who prefer to use prepared tables with corrections for changes in temperatures will find their use in estimating prices explained in the next section (pp. 41-44). The reason for using the logarithms of receipts and of estimated purchasing power, i.e., log A and log X, is that changes in receipts in terms of absolute numbers and changes in purchasing power in terms of absolute numbers do not show a straight - line relationship, but changes in receipts expressed as ratios and changes in purchasing power expressed as ratios do show a straight- line relationship, and are therefore more easily handled in a corre- lation analysis in terms of logarithms than in terms of the original numbers. The following illustrations show the use of this equation. Illustration 1. On the basis of the above equation, let us estimate the price for a certain week in 1927, assuming a definite set of con- ditions. Take, for example, the first week, and assume that shipments from Imperial Valley and diversions indicate average receipts of ten carlots a day for the first week, and an average maximum temperature lagged three days (from Thursday of the previous week to Wednesday of the week of which the receipts are based) of 70 degrees Fahrenheit. The values of the factors in the equation would then be A = 10 and log A = 1 b = — 2.6. Since the average temperature for the first week of receipts was 72.6° F 6 the temperature of 70° is —2.6° deviation from the average. = 1, the first week of receipts. Substituting these numbers in the equation : log X= 0.9686 — 0.3036 log A + 0.00716 — 0.0210C gives the following : log X = 0.9686— (0.3036X1) + (0.0071X— 2.6) — (0.0210X1). Solving, we obtain : log X = 0.9686 — 0.3036 — 0.0185 — 0.0210 log X = 0.6255. Looking up the anti-log of 0.6255, we obtain X = $4.22. This represents the purchasing power of a standard crate. Assuming now 6 The average maximum temperatures in degrees Fahrenheit were : 1st week, 6 " 79.0. 40 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION an index number of 155 for 1927 would give a price, expressed in 1927 dollars, of $4.22 multiplied by 1.55, or $6.54. 7 Thus if the same relationships prevail between price and receipts, temperature, and time, in 1927, as have prevailed from 1921 to 1926, the most probable price for a standard crate of cantaloupes (salmon tint 45 's), for the conditions assumed, would be $6.54. The above method was used in calculating the estimated prices shown in table 16 and in figure 8. Illustration 2. Let us assume that shipments indicate average daily receipts at New York during the fifth week (Monday to Satur- day) of 70 carlots and that maximum temperatures from the previous Thursday to Wednesday of this week will average 85 degrees Fahren- heit. What will be the most probable price based on the relationship of the previous six years? Using the same symbols as in the previous illustration we have : A = 10 carlots and log A = 1.8451 & = 85°F — 77.6°, 8 or 7.4° C = 5. Substituting in the equation : log X = 0.9686 — 0.3036 log A + 0.00715 — 0.0210C we obtain log T= 0.9686— (0.3036 X 1.8451) + (0.0071 X 7.4) — (0.0210 X 5) log X = 0.9686 — 0.5602 + 0.0525 — 0.1050 log X = 0.3559. 7 The following change in the method of applying the above equation which has been suggested by Professor C. M. Titus of this Experiment Station, combines the adjustment for the change in the index number with the regression equation, and hence reduces by one the number of calculations necessary in arriving at the estimated price. Since log X multiplied by 1.55 equals the estimated price in 1927 dollars, we may add the logarithm of 1.55 to the right side of the equation. The equation then will read: log of estimated price in 1927 dollars = 0.9686 + log 1.55 — 0.3036 log A + 0.0071& — 0.0210C log of estimated price in 1927 dollars t= 0.9686 + 0.1903 — 0.3036 log A + 0.0071& — 0.0210C log of estimated price in 1927 dollars = 1.1589 — 0.3036 log A + 0.00716 — 0.0210C. Applying this equation to the conditions assumed in illustration 1, we secure log of estimated price in 1927 dollars = 1.1589— (0.3036 X 1) + (0.0071 X —2.6) — (0.0210 X 1) = 1.1589 — 0.3036 — 0.0185 — 0.0210 = 0.8158. Looking up the anti-log of 0.8158 we obtain $6.54 — the estimated price per standard crate expressed in 1927 dollars. 8 The average maximum temperature for the fifth week as given in footnote 6. BUL. 419] ECONOMIC ASPECTS OP THE CANTALOUPE INDUSTRY 41 Looking up the anti-log of 0.3559, we obtain ~X= $2.27. This represents the purchasing power of a standard crate. Assuming now an index number of 155 for 1927, as in the previous illustration, we arrive at the price, expressed in 1927 dollars, of $2.27 multiplied by 1.55, or $3.52. Illustration 3. It would be of interest to know what the estimated price would be for the same week and temperatures, but with an average of 60-carlot receipts daily instead of 70. With this change our equation would be log ~X = 0.9686— (0.3036 X 1.7782) + (0.0071 X 7.4) — (0.0210 X 5) log X = 0.9686 — 0.5399 + 0.0525 — 0.1050 log X = 0.3762. Looking up the anti-log of 0.3762, we obtain X = $2.38. Assuming an index number of 155 for 1927 would give a price, expressed in 1927 dollars, of $2.38 multiplied by 1.55, or $3.69. This price is $0.17 higher per standard crate than would be expected with receipts averaging 70 carlots daily, instead of 60. Estimating Prices by Means of Tables and Charts with Corrections for Changes in Temperatures. — Because of the difficulty that may be encountered by some readers in using the equation for estimating prices as described in the preceding sections, tables 17 and 18 and figure 9 are added. These give a number of estimated prices based upon the relationships expressed in the equation. It can be readily seen that with receipts in New York varying from 1 to 80 carlots daily during a season, average weekly maximum temperatures rang- ing from 62° to 88°, and weeks of each season numbering from 1 to 6, thousands of combinations would be possible. A complete table show- ing the estimated prices for all of these combinations would not be so usable as a table giving estimated prices for a limited number of combinations of these factors whose relationships to actual prices during the past six years have been calculated. Table 17 gives the estimated prices which we would expect in 1927, assuming a price level of 155, and the receipts and week of the season as indicated. Temperatures are assumed to be the average in this table. Thus with average receipts of 5 carlots during the first week and average tem- peratures, the expected price of a standard crate is $8.43. With average receipts of 15 carlots daily during the second week and average temperatures, a price of $5.75 would be expected. Figure 9 illustrates table 17 and also shows the gradations in estimated prices for receipts between 5, 10, 15, etc., carlots. 42 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION The probable effect of temperature on price can be calculated by multiplying the prices given in table 17 (or as read from the curves in figure 9) by the appropriate correction factor in table 18 if the Probable Prices of Imperial Valley Cantaloupes at New York for Successive Weeks, and Varying Carlot Eeceipts Vof/ors Per- Cra-fre grts-f Week Week- — > 3r< i Week^^ 4-H Week- — y Si-/} Week — -^ Week — * O /O ZO JO 40 50 -fe Fig. 9. — The average net effect of the week of the season and of receipts on prices from 1921 to 1926, assuming average temperatures. The average prices of a crate for the second week are 4.5 per cent lower than the prices prevailing for the same volume of receipts the first week. The third week's prices are 4.5 per cent below those of the second week for the same volume of receipts, and so on for the rest of the six weeks. TABLE 17 Estimated Prices* for 1927 by Specific Weeks and Carlot Eeceipts, Assuming Average Temperatures for Each Week Receipts carlots per day 1st week 2nd week 3rd week 4th week 5th week 6th week 5 $8.43 10 6.83 $6.52 15 6.04 5.75 20 5.53 5.27 $5.03 30 4.67 4.44 $4.24 40 4.27 4.07 3.88 $3.70 $3.52 50 3.81 3.63 3.46 3.29 60 3.61 3.43 3.27 3.11 70 3.14 2.97 80 2.85 * These prices were obtained from the following formula: log X =+l. 1589-0.3036 log A+0.0071 b -0.0210C These prices can be corrected for deviations from the average temperatures by using the correction factor given jn table 18. BUL. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 43 temperature is above the average, or dividing by the appropriate correction factor if the temperature is below the average. To illustrate the use of these tables let us assume the same conditions as in the preceding section. TABLE 18 Data for Correcting Price Estimates Given in Table 17 for Deviations from Average Temperatures Deviations from average maximum temperatures* in degrees Fahrenheit Temperature correction factor 1.000 1 1 017 2 1 033 3 1 050 4 1 068 5 1 085 6 1 103 7 1 121 8 1 140 9 1 159 10 1 178 11 1 197 12 1 217 * Average maximum temperatures in degrees Fahrenheit for the seasons of 1921-1926 were: 1st week 72.6°, 2nd week 74.8°, 3rd week 77.5°, 4th week 77.8°, 5th week 77.6°, 6th week 79.0°. Illustration 1. On the basis of the above tables let us estimate the price for a certain week in 1927, assuming a definite set of conditions. Take, for example, the first week and assume that shipments from Imperial Valley and diversions indicate average receipts of 10 carlots daily for the first week, and an average maximum temperature lagged three days (from Thursday of the previous week to Wednesday of the week in which the receipts are based) of 70 degrees. Referring to table 17, we note that with average temperatures, daily receipts of 10 carlots the first week would be expected to result in a price of $6.83 for a standard crate. However, the temperature assumed is 70 degrees Fahrenheit, while the average temperature for the first week has been 72.6 degrees (see footnote table 18). Hence, we must calculate the probable effect of the low temperature by dividing $6.83 by the proper correction factor. The nearest whole number for the deviation of 2.6 degrees from the average is 3 degrees (table 18), for which the correction factor given is 1.050. This factor, divided into $6.83, gives an estimated price of $6.50. If one wishes to go to the trouble of getting the correction factor for 2.6 degrees instead of taking the nearest whole number, one can 44 UNIVERSITY OF CALIFORNIA EXPERIMENT STATION get the more exact correction factor by interpolation. That is, the correction factor for 2 degrees is 1.033, and for 3 degrees, 1.050. Taking six-tenths of the difference between these correction factors (0.6 X 0.017) equals 0.010. Adding 0.010 to 1.033 equals 1.043— the correct factor for a deviation of 2.6 degrees from the average. This divided into $6.83 gives $6.54 as the estimated price of a standard crate for the conditions assumed, which is the same as the estimated price calculated by means of the equation in illustration 1 of the previous section. Of course, for practical purposes there is no point in adjusting for fractional differences in temperature. The only purpose in calculating it here is to show that the estimated prices calculated from the equation in the previous sections and from tables 17 and 18 of this section yield identical results. Illustration 2. Let us assume the same conditions as in illustra- tion 2 of the previous section, viz., daily receipts at New York of 70 carlots for the fifth week, and average maximum temperatures from the previous Thursday to the Wednesday of this week of 85 degrees. For average temperatures the most probable price would be $3.14 (table 17). However, the temperature assumed here is 7.4 degrees (85 — 77.6) above the average. The correction factor for 7 degrees (table 18) is 1.121. Multiplying this factor by $3.14 gives $3.52 as the estimated price of a standard crate. The estimated price secured by the equation in the previous section, illustration 2, is also $3.52. It should be noted again at this point that in correcting for devia- tions above the average temperature the correction factor (table 18) is multiplied by the estimates shown in table 17, but in correcting for deviations below the average temperature, the correction factor is divided into the estimates shown in table 17. ACKNOWLEDGMENTS Acknowledgments are due W. P. Hedden, Research Agent, Port of New York Authority and U. S. Bureau of Agricultural Economics, for valuable suggestions in planning this investigation. Also to C, EL Schultz, local representative of the U. S. . Bureau of Agricultural Economics, and to O. W. Schleussner, of the American Fruit Growers, Inc., for reading the second draft of the manuscript and offering constructive criticism. C. M. Titus, Associate Statistician of this Experiment Station, read the second draft of the manuscript and carefully checked the statistical methods used. Bul. 419] ECONOMIC ASPECTS OF THE CANTALOUPE INDUSTRY 45 LITERATURE CITED (1) SCHULTZ, C. E., AND GEO. E. PRINCE. 1926 (May). Summary of western cantaloupe deals, season 1925. U.S. Bur. of Agr. Econ. Mimeo. Report, 1-56. (2) Lamb, R. H. 1924 (Aug.). Colorado cantaloupe deal, season 1923. U.S. Bur. of Agr. Econ. Mimeo. Report, 1-18. (3) Schultz, C. E. 1926 (July 2). Preliminary review of the 1926 Imperial Valley cantaloupe deal. U. S. Bur. of Agr. Econ. Mimeo. Report, 1-8. (4) SCHLEUSSNER, O. W., AND C. W. KlTCHEN. 1916. Marketing and distribution of western muskmelons in 1915. U. S. Dept. of Agr. Bul. 401:1-38. (5) Mills, F. C. 1924. Statistical methods. 1-604. Henry Holt and Co., New York. (6) Miller, O. D. 1923. Review of the 1923 season for California cantaloupes on the New York market. U. S. Bur. of Agr. Econ. Mimeo. Report, 1-3. (7) Callanan, V. D., and S. W. Russell. 1924. Review of the 1924 season for California contaloupes on the New York market. U. S. Bur. of Agr. Econ. Mimeo. Report, 1-7. (8) Warren, G. F. 1923. Prices of farm products in New York. Cornell Univ. Agr. Exp. Sta. Bul. 416:1-63. ( 9 ) Dept. of Agr. Econ. and Farm Management, New York State College of Agr., Cornell Univ. 1926 (May 15). Farm Economics 34:441-456. do) Callanan, V. D., and C. E. Schultz. 1923. Imperial Valley cantaloupe deal, season 1923. Mimeo. Report, 1-24. (id Wallace, H. A., and Geo. W. Snedecor. 1925. Correlation and machine calculation. Iowa State College of Agr. and Mechanic Arts Official Publication 23 35 :l-47. STATION PUBLICATIONS AVAILABLE FOR FREE DISTRIBUTION No. 253. Irrigation and Soil Conditions in the Sierra Nevada Foothills, California. 261. Melaxuma of the Walnut, "Juglans regia." 262. Citrus Diseases of Florida and Cuba Compared with Those of California. 263. Size Grades for Ripe Olives. 268. Growing and Grafting Olive Seedlings. 273. Preliminary Report on Kearney Vine- yard Experimental Drain. 275. The Cultivation of Belladonna in California. 276. The Pomegranate. 277. Sudan Grass. 278. Grain Sorghums. 279. Irrigation of Rice in California. 283. The Olive Insects of California. 294. Bean Culture in California. 304. A Study of the Effects of Freezes on Citrus in California. 310. Plum Pollination. 312. Mariout Barley. 313. Pruning Young Deciduous Fruit Trees. 319. Caprifigs and Caprification. 324. Storage of Perishable Fruit at Freez- ing Temperatures. 325. Rice Irrigation Measurements and Experiments in Sacramento Valley, 1914-1919. 328. Prune Growing in California. 331. Phylloxera-Resistant Stocks. 335. Cocoanut Meal as a Feed for Dairy Cows and Other Livestock. 339. The Relative Cost of Making Logs from Small and Large Timber. 340. Control of the Pocket Gopher in California. 343. Cheese Pests and Their Control. 344. Cold Storage as an Aid to the Mar- keting of Plums. 346. Almond Pollination. 347. The Control of Red Spiders in Decid- uous Orchards. 348. Pruning Young Olive Trees. 349. A Study of Sidedraft and Tractor Hitches. 350. Agriculture in Cut-over Redwood Lands. 352. Further Experiments in Plum Pollina- tion. 353. Bovine Infectious Abortion. 354. Results of Rice Experiments in 1922. 357. A Self-mixing Dusting Machine for Applying Dry Insecticides and Fungicides. 358. Black Measles, Water Berries, and Related Vine Troubles. 361. Preliminary Yield Tables for Second Growth Redwood. 362. Dust and the Tractor Engine. 363. The Pruning of Citrus Trees in Cali- fornia. 364. Fungicidal Dusts for the Control of Bunt. 365. Avocado Culture in California. 366. Turkish Tobacco Culture, Curing and Marketing. 367. Methods of Harvesting and Irrigation in Relation of Mouldy Walnuts. 368. Bacterial Decomposition of Olives dur- ing Pickling. 369. Comparison of Woods for Butter Boxes. BULLETINS No. 370. 371. 37'. 373. 374. 375. 376. 377. 379. 380. 381. 382. 383. 385. 386. 387. 388. 389. 390. 391. 392. 393. 394. 395. 396. 397. 398. 399. 400. 401. 402. 403. 404. 405. 406. 407. 408. 409. 410. 411. 412. 413. 414. Browning of Yellow Newtown Apples. The Relative Cost of Yarding Small and Large Timber. The Cost of Producing Market Milk and Butterfat on 246 California Dairies. Pear Pollination. A Survey of Orchard Practices in the Citrus Industry of Southern Cali- fornia. Results of Rice Experiments at Cor- tena, 1923. Sun-Drying and Dehydration of Wal- nuts. The Cold Storage of Pears. Walnut Culture in California. Growth of Eucalyptus in California Plantations. Growing and Handling Asparagus Crowns. Pumping for Drainage in the San Joaquin Valley, California. Monilia Blossom Blight (Brown Rot) of Apricot. Pollination of the Sweet Cherry. Pruning Bearing Deciduous Fruit Trees. Fig Smut. The Principles and Practice of Sun- drying Fruit. Berseem or Egyptian Clover. Harvesting and Packing Grapes in California. Machines for Coating Seed Wheat with Copper Carbonate Dust. Fruit Juice Concentrates. Crop Sequences at Davis. Cereal Hay Production in California. Feeding Trials with Cereal Hay. Bark Diseases of Citrus Trees. The Mat Bean (Phaseolus aconitifo- lius). Manufacture of Roquefort Type Cheese from Goat's Milk. Orchard Heating in California. The Blackberry Mite, the Cause of Redberry Disease of the Himalaya Blackberry, and its Control. The Utilization of Surplus Plums. Cost of Work Horses on California Farms. The Codling Moth in Walnuts. Farm-Accounting Associations. The Dehydration of Prunes. Citrus Culture in Central California. Stationary Spray Plants in California. Yield, Stand and Volume Tables for White Fir in the California Pine Region. Alternaria Rot of Lemons. The Digestibility of Certain Fruit By- products as Determined for Rumi- nants. Factors Affecting the Quality of Fresh Asparagus after it is Harvested. Paradichlorobenzene as a Soil Fumi- gant. A Study of the Relative Values of Cer- tain Root Crops and Salmon Oil as Sources of Vitamin A for Poultry. The California Poultry Industry; a Statistical Study. Planting and Thinning Distances for Deciduous Fruit Trees. the the No. 87. Alfalfa. 117. The Selection and Cost of a Small Pumping Plant. 127. House Fumigation. 129. The Control of Citrus Insects. 136. Melilotus indica as a Green-Manure Crop for California. 144. Oidium or Powdery Mildew of Vine. 157. Control of the Pear Scab. 160. Lettuce Growing in California. 164. Small Fruit Culture in California. 166. The County Farm Bureau. 170. Fertilizing California Soils for 1918 Crop. 173. The Construction of the Wood-Hoop Silo. 178. The Packing of Apples in California. 179. Factors of Importance in Producing Milk of Low Bacterial Count. 190. Agriculture Clubs in California. 199. Onion Growing in California. 202. County Organizations for Rural Fire Control. 203. Peat as a Manure Substitute. 209. The Function of the Farm Bureau. 210. Suggestions to the Settler in California. 212. Salvaging Rain-Damaged Prunes. 215. Feeding Dairy Cows in California. 217. Methods for Marketing Vegetables in California. 220. Unfermented Fruit Juices. 228. Vineyard Irrigation in Arid Climates. 230. Testing Milk, Cream, and Skim Milk for Butterfat. 231. The Home Vineyard. 232. Harvesting and Handling California Cherries for Eastern Shipment. 234. Winter Injury to Young Walnut Trees during 1921-22. 235. Soil Analysis and Soil and Plant Inter-relations. 236. The Common Hawks and Owls of California from the Standpoint of the Rancher. 237. Directions for the Tanning and Dress- ing of Furs. 238. The Apricot in California. 239. Harvesting and Handling Apricots and Plums for Eastern Shipment. 240. Harvesting and Handling Pears for Eastern Shipment. 241. Harvesting and Handling Peaches for Eastern Shipment. 243. Marmalade Juice and Jelly Juice from Citrus Fruits. 244. Central Wire Bracing for Fruit Trees. 245. Vine Pruning Systems. 247. Colonization and Rural Development. 248. Some Common Errors in Vine Prun- ing and Their Remedies. 249. Replacing Missing Vines. 250. Measurement of Irrigation Water on the Farm. 252. Supports for Vines. 253. Vineyard Plans. 254. The Use of Artificial Light to Increase Winter Egg Production. CIRCULARS No. 255. 256. 257. 258. 259. 261. 262. 263. 264. 265. 266. 267. 269. 270. 272. 273. 274. 276. 277. 278. 279. 281. 282. 283. 284. 285. 286. 287. 288. 289. 290. 291. 292. 293. 294. 295. 296. 298. 299. 300. 301. 302. 303. 304. 305. 306. 307. 308. Leguminous Plants as Organic Fertil- izer in California Agriculture. The Control of Wild Morning Glory. The Small-Seeded Horse Bean. Thinning Deciduous Fruits. Pear By-products. Sewing Grain Sacks. Cabbage Growing in California. Tomato Production in California. Preliminary Essentials to Bovine Tuberculosis Control. Plant Disease and Pest Control. Analyzing the Citrus Orchard by Means of Simple Tree Records. The Tendency of Tractors to Rise in Front; Causes and Remedies. An Orchard Brush Burner. A Farm Septic Tank. California Farm Tenancy and Methods of Leasing. Saving the Gophered Citrus Tree. Fusarium Wilt of Tomato and its Con- trol by Means of Resistant Varieties. Home Canning. Head, Cane, and Cordon Pruning of Vines. Olive Pickling in Mediterranean Coun- tries. The Preparation and Refining of Olive Oil in Southern Europe. The Results of a Survey to Determine the Cost of Producing Beef in Cali- fornia. Prevention of Insect Attack on Stored Grain. Fertilizing Citrus Trees in California. The Almond in California. Sweet Potato Production in California. Milk Houses for California Dairies. Potato Production in California. Phylloxera Resistant Vineyards. Oak Fungus in Orchard Trees. The Tangier Pea. Blackhead and Other Causes of Loss of Turkeys in California. Alkali Soils. The Basis of Grape Standardization. Propagation of Deciduous Fruits. The Growing and Handling of Head Lettuce in California. Control of the California Ground Squirrel. The Possibilities and Limitations of Cooperative Marketing. Poultry Breeding Records. Coccidiosis of Chickens. Buckeye Poisoning of the Honey Bee. The Sugar Beet in California. A Promising Remedy for Black Measles of the Vine. Drainage on the Farm. Liming the Soil. A General Purpose Soil Auger and its Use on the Farm. American Foulbrood and its Control. Cantaloupe Production in California. The publications listed above may be had by addressing College of Agriculture, University of California, Berkeley, California. 12m-2,'27