SB 211 UC-NRLF A Study, by the Crop Survey Method, of "Factors Influeneing the Yield -•;, of Potatoes A THESIS PRESENTED TO THE FACULTY OF THE GRADUATE SCHOOL OF CORNELL UNIVERSITY FOR THE DEGREE OF 'DOCTOR OF PHILOSOPHY BY EARLE VOLCARTIHARDENBURG Published as Memoir 57 Cornell University Agricultural Experiment Station, June 1922. A Study, by the Crop Survey Method, of Factors Influencing the Yield of Potatoes A THESIS PRESENTED TO THE FACULTY OF THE GRADUATE SCHOOL OF CORNELL UNIVERSITY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY BY EARLE VOLCARTJHARDENBURG Published as Memoir 57 Cornell University Agricultural Experiment Station, June 1922. 517499 CONTENTS ' '• ' 't ' , '" > ' ' « • • PAGE The crop survey as a method of research ....'.','. .'.'. . ; , ... , . .'./. • 'J; • .,.;. \ . '. -• 1145 Biometry as applied to crop-survey data '..'.' 1147 The taking cf records ' 1147 Description of regions surveyed 1148 Long Island 1149 Steuben County ..:..- 1153 Monroe County 1154 Franklin and Clinton Counties 1155 Method of studying survey data 1156 The study of factors 1157 Climate 1157 Elevation 1159 Crop rotation 1162 Value of land. . . . '. 1166 Soil ' 1169 Plowing , 1171 » Time of plowing 1171 Depth of plowing .• 1173 Manure and fertilizer 1177 Home-mixed fertilizers 1179 Quantity and value of fertilizer 1180 Manure and fertilizer used in the four regions 1182 Analysis of fertilizer 1188 Method of applying fertilizer 1191 Use and influence of lime on potato land 1196 Adaptation and yield of varieties 1197 Source of seed .' 1205 Sun-sprouting of seed 1208 Chemical treatment of seed 1210 Interval between cutting and planting. 1211 Dusting cut seed 1214 Type of seed 1215 Large as compared with small tubers for seed 1215 Whole as compared with cut seed 1216 Large as compared with small seed pieces 1217 Number of eyes 1220 Types of seed used in the four regions surveyed . . 1221 Relation of amount of seed to yield 1224 Date of planting 1231 Hand as compared with machine planting t 1232 Checkrow as compared with drill planting 1234 Depth of planting 1237 Depth of cultivation ; 1242 Ridge as compared with level culture 1244 Frequency of cultivation .' . 1246 Spraying . 1255 Relation of date of harvest to yield 1263 Method of harvesting in the four regions surveyed 1265 Types of storage in the four regions surveyed 1269 Length of storage period 1269 Summary 1270 Conclusions 1272 Author's acknowledgment 1273 Bibliography 1274 Survey blank 1280 1139 A STUDY, BY THE CROP SURVEY METHOD, OF FACTORS INFLUENCING THE YIELD OF POTATOES A STUDY, BY THE CROP SURVEy METHOD, 'OF; FACTORS INFLUENCING THE YIELD OF POTATOES1" EARLE V. HARDENBURG Almost from the date of their establishment, practically all state and federal experiment stations in this country, as well as many foreign stations, have tested, by diverse methods, the relative influences of factors affecting the yield of potatoes. A review of the abundant literature of the subject shows that a majority of these tests concern the influence of seed and fertilizers on yield. This fact, further substantiated by the results of the study herein reported, indicates that, with the exception of climate and soil, seed and fertilizers are the most vital factors affect- ing yield. Because of the widely differing environmental conditions under which the tests have been conducted, it is possible in only a limited degree to draw definite conclusions from a summary of the results. Furthermore, a large part of the literature fails to supply much detailed information as to the methods used in the experiments, and gives little if any considera- tion to factors affecting yield other than the one principally concerned in the respective tests. This means that most of the evidence available to date is of only limited application. A comparison of the conclusions reached and the recommendations made by experiment stations, with those warranted by actual practice as found on farms in a potato-growing region, is therefore of considerable value. Such a comparison is, to some extent, made possible by the use of the survey method of collecting and studying data on the influ- ences affecting the yield of potatoes. The survey method has accordingly been applied to the study of such factors in several potato sections of New York, and the results are herein compared with those obtained experimentally. As an additional check on the conclusions drawn, the biometrical method as applied by Rietz and Smith (1910)2 has also been used in studying those factors which, according to the sur- vey method, appear to affect the yield to the greatest extent. The survey method of studying crop production, wholly aside from the agricultural methods involved, has proved to be a most valuable means of determining the actual practice thruout the State, and has aided in correcting many false ideas of long standing concerning cultural methods used with this crop. The collection of data was begun in the summer of 1913 and continued thru the summer of 1914. In 1913, 330 records of the 1912 potato crop were taken from as many potato farms on Long Island, and 360 records 1 Also presented to the Faculty of the Graduate School of Cornell University, February, 1919, as a major thesis in partial fulfillment of the requirements for the degree of doctor of philosophy. ' 2 Dates in parenthesis refer to Bibliography, page 1274. 1143 1144 V. HAKDENBURG of the riame, year V: crop were obtained from that number, of farms in northern Steuberi County. In 1914, 300 records were similarly taken for the 1913 crop in Monroe County, and 300 in Franklin and Clinton Counties combined. This gives a total of 1290 records for the crops of 1912 and 1913. Each record was in the form of a filled-out survey blank, a sample of which is included at the end of this paper, and was as complete as possible in the details listed. Because of the similarity of regional conditions and of cultural practices, the counties surveyed were studied FlG. 125. REGIONS INCLUDED IN THE POTATO SURVEY asjfour distinct sections, as follows: (1) Long Island, including the potato- growing areas of Suffolk and Nassau Counties; (2) Steuben County; (3) Monroe County; (4) Franklin and Clinton Counties. The location of these areas is shown in figure 125. These regions were selected, not because they include the counties of highest total production, but because they represent typical and distinct centers of potato production in the State. The importance of potato production in a region is probably best indicated by figures showing the percentage of total crop acres devoted to this crop and the average potato acreage per farm. A summary of A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1145 the scope of the survey and the status of the industry during the years 1912 and 1913 is given in table 1. Of the four regions surveyed, the potato crop is regarded as of most importance on Long Island and of least importance in Franklin and Clinton Counties. TABLE 1. SUMMARY OF THE FOUR REGIONS SURVEYED Region Year Number of fecords Acreage surveyed Average size of farm surveyed (acres) Per cent of total acreage in potatoes, for farms sur- veyed Per cent of crop acreage in potatoes, for farms sur- veyed Per , cent of crop" acreage in potatoes, for the countv (1909 census) Average potato acreage per farm sur- veyed Average yield per acre for farms sur- veved (bushels) Long Island, in- cluding parts of Suffolk and Nassau Coun- ties 1912 330 8,188 16 65 0 37 44 23 0 24.8 175.5 Steuben County. . Monroe County . . Franklin and Clin- ton Counties.. . 1912 1913 1913 360 300 300 5,301.10 3,728.25 2,160.00 145.8 112.1 169.5 10 11 4 18 15 10 8.3 8.0 5.7 14.7 12.4 7.2 136.4 126.2 179.3 THE CROP SURVEY AS A METHOD OF RESEARCH From its inception in this country, agricultural teaching has depended largely on textbooks, collateral references, and the published results of experiments. There is still a considerable lack of practical information which can be supplied only by protracted experimentation or by the study of large numbers of survey records in the regions concerned. Fre- quently problems arise which local experiments fail to solve because of the impossibility of handling the work on a sufficiently extensive scale. Large numbers of records might very often be the means of discovering the common causal factor prevailing thruout a region, thus furnishing the solution of the problem or at least a working basis for its solution. A typical illustration of this is furnished in the investigations on pecan rosette by McMurran (1919). Pathologists had previously been unable to account for the cause or to recommend measures for the control of this disease, which was so prevalent thruout the pecan orchards of the Southern States. McMurran, by taking records of many orchards in the various pecan regions of the South, discovered that the disease was almost entirely absent in the orchards of the rich river bottom- ands, and from this observation he deduced that the cause of the disease lay in certain soil deficiencies. The farm-crops survey aims first of all to search out the actual facts concerned in the production of a given crop in a given area. This informa- tion, obtained in sufficient quantity, may then be regarded as statistics 1146 EARLE V. HARDENBURG from which to determine the most beneficial influences and practices. The survey idea was first launched in New York by Dr. L. H. Bailey, under whose direction horticultural studies were made thruout the State. In 1903 Professor John Craig started an orchard survey campaign in west- ern New York. Such of these surveys as were completed have been published as Cornell bulletins (Warren, 1905, a and b; Cummings, 1909; Martin, 1911). Under the direction of Dr. G. F. Warren, the survey idea was extended to include the farm as a whole, with the result that whole farming areas, with the farm as a unit, have been studied in what are called farm-management surveys. The results of such studies have also been published as Cornell bulletins (Warren and Livermore, 1911; Thompson, 1915). Montgomery (1913), in discussing crop surveys, states that their primary function is to determine how to grow the crop, while farm-management surveys aim to determine when to grow the crop. Warren (1914) attests the value of agricultural surveys by saying that there are many kinds of agricultural information that can be found only by survey methods, since the conditions in question exist only on the farms. He states further that agricultural knowledge, to be of most value, should be the result of both survey studies and experimental tests. The accuracy of survey methods depends very largely on such factors as the personality of the man procuring the records, the manner in which questions are asked, the number of records obtained for each region studied, the unit used as a basis in the study of a factor, accuracy in tabulation, and the final interpretation of results. The more extensive the record to be obtained, the greater is the number of records necessary for final accuracy. The principal faults in much of the survey work to date lie in the attempt to include too much detail and in the use of too- few records. Warren (1914) is of the opinion that ordinarily 1000 records should be used, tho 500 may be enough in some cases. However, the necessity of such large numbers depends somewhat upon the scope of the survey. By the law of averages, large numbers tend to eliminate individual errors. Spillman (1917) has said that the accuracy of any average depends on three things: first, on freedom from bias; secondly, on the number of items from which an average is obtained; and thirdly, on the accuracy of the individual, items averaged. Considering the limitations of much of the experimental evidence to date, large numbers of survey records are undoubtedly pro- ductive of as nearly accurate results as are obtained by experimental work. As . emphasized by Warren (1914), the region surveyed should be an agricultural, not a political, unit. Furthermore, the records should be taken only during a normal year unless records are to be obtained for con- secutive years. Unfortunately for this study, the year 1912 was at first drier than normal, but the abundant rain which fell late in the growing season caused some blight rot; 1913, however, was a more nearly normal year. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1147 BIOMETRY AS APPLIED TO CROP-SURVEY DATA Biometry as a science is beginning to have wide application, wherever sufficient data make its application possible, in the solution of problems involving the study of the interrelation of factors or the study of cause and effect. Until the present time, biometry has been used mainly only in the study of inheritance and in the correlation of characters in large popu- lations of plants and animals. Its use has been thus limited because only in such studies have conditions been so controlled that none but the factor or factors under observation could affect the results, and because it has been possible to use large numbers of individuals for such investigations. Biometry should have a place in the study of crop-survey data wherever large numbers of records are involved, in order that the coefficient of correlation may serve as a check on the conclusions otherwise drawn and that it may furnish, thru its frequency table, a description of the prevailing practice in the region in question. Tolley (1917) states that the coefficient of net correlation affords a good means of determining the net effect of each of several factors bearing on a result, or of eliminating the effect of other factors when it is desired to find the true relationship between any two. Applying biometrical methods to farm-survey data on fattening baby beef, Tolley has shown how the 'gross apparent correlation between any two or more factors may be substituted in a derived formula and the net correlation of any two factors thereby deduced. -r^'l A biometrical analysis of some of the more influential factors involved in this study has" been made, altho, owing to the relatively large numbers of records used in each study, only the gross correlation has been computed. Aside from the actual significance of the coefficients obtained, much information of descriptive value relative to the frequency of a given practice may be found in the frequency distribution tables. One of the chief functions of biometry is description. It affords a means of classifying a group of individuals not possible by any other means. THE TAKING OF RECORDS Five men constituted the party employed in the taking of records in 1913. This made it possible for four of the party to travel thru the potato regions in pairs while the fifth man copied and checked each day's records. In this way, any discrepancies in the records could be checked up by a return visit to the grower or by discussion within the party. The data on the 1913 crop were taken in 1914 by two ,men. As previously noted (Spillman, 1917), the value and accuracy of survey data depend largely on freedom from bias. This may well apply to thje selection of farms to be observed. Therefore it was decided that for these surveys the only limitation in the selection of a farm was to be in. the acreage of the crop produced the previous year. This limitation was set 1148 EARLE V. HARDENBURG at a minimum of 5 acres, tho a very few records were taken on farms having a production area of only 4 acres. The reason for the establish- ment of this minimum limit lies in the assumption that growers of acreages smaller than 5 are probably not growing potatoes in a manner comparable to the average of the region. The data on cost of production, tho obtained at the same time and indicated on the survey blank, are not a part of this study. The subject of cost has been studied by Fox (1919), formerly of the Department of Farm Management at Cornell University. DESCRIPTION OF REGIONS SURVEYED For a better understanding of the environmental conditions under which the potato crop was produced, a brief description of climate, soil, topography, elevation, length of growing season, market facilities, type of farming, land values, and status of potato production, is given for each of the four regions surveyed. Unfortunately, of the regions concerned, only Monroe and Clinton Counties have been soil-surveyed by the United States Department of Agriculture. More detailed knowledge as to these environmental influences may be obtained from figures 126 to 129. FlG. 126. ELEVATIONS OF THE REGIONS SURVEYED A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1149 LONG ISLAND Most of the potato crop in Suffolk County is grown east of Riverhead on both the north and the south shores of Long Island. The Long Island Railroad furnishes the transportation facilities for practically all of the surplus crop of this region. Most of the roads are improved to a high degree. Thus the time required for shipments to reach New York City need not be over one day and no delay is necessitated by transfers to other railroads. FlG. 127. GROWING-SEASON RAINFALL (iN INCHES) IN REGIONS SURVEYED Nearly all of the crop in Nassau County is grown north of a line drawn east and west thru the central part of the county. Most of the surplus crop of this county is transported directly, in heavy wagons and motor trucks, to the Wallabout and Harlem Markets in Brooklyn. The greater part of Long Island is of marine deposit formation, the elevation ranging from a point at about sea level, in the Hampton section, to an altitude of nearly 300 feet in some places on the north shore. The average elevation of the potato fields surveyed was 65.5 feet. Due to the low elevation of the south shore, the crop is exposed to heavy sea fogs which make conditions favorable to the development of late blight. The 1150 EARLE V. HARDENBURG topography is in 'general fairly level, tho the slightly rolling lands along the north shore gradually rise until they merge into prominent hills along the Sound. The potato sections of Long Island show an average growing-season ra'in- fall of from 16 to 20 inches, which is somewhat higher than that of most of the potato sections of New York. Rainfall seldom limits production here. The tempering influence of the Atlantic Ocean affords a growing season of approximately 200 days between killing frosts, which is greater than that FlG. 128. LENGTH OF GROWING SEASON (iN DAYS) IN REGIONS SURVEYED of any other section of New York. The growing season on Long Island is fully a month earlier than that in the other three regions under dis- cussion. The soil of most of the potato-growing areas of Long Island is of a sandy texture, topped by silty loam in layers of varying thickness. This is counter to a rather common impression that the Long Island crop is pro- duced in sandy soil. The greater part of .the central section of the island does consist of sand, and this supports little vegetation aside from scrub oak and pine. That the potatoes are grown mainly on the Sassafras series of soil is shown in figure 129. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1151 1152 EARLE V. HARDENBURG The importance of the crop on Long Island is shown by the fact that no regular system of crop rotation is practiced, potatoes being grown for several successive years on the same land. In order to maintain the humus content, cover crops of rye are turned under each spring. The commonest practice is two to four years of potatoes, the land being cover- cropped to rye over winter. Along the north shore, where a rotation is sometimes used, wheat seeded to clover and timothy follows potatoes, the hay being grown from one to two years before the sod is plowed for corn, cabbage, and cauliflower. Potatoes then follow these cultivated FlG. 130. HARVESTING IRISH COBBLERS IN NASSAU COUNTY IN JULY The large immature vines should be noted crops. Wheat and hay are the principal rotation crops on the southern shore. Much double-cropping is practiced in Nassau County, the early potatoes being harvested in July and the second crop in late August and early Sep1 tember. Land producing a first crop of potatoes is commonly planted to turnips, beets, carrots, or other root crops, or is set to cabbage for the fall market. Rye is used here also as a cover crop. A field in which Cobblers were harvested one day and turnips were planted the next day, is shown in figure 130. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1153 Land values are higher on Long Island than in the other potato sections, partly because much of the land in Nassau County is held for real-estate purposes and partly because of its geographical advantages and adapt- ability for potato production. The values range from $100 an acre in Suffolk County, to $1000 an acre, real-estate value, in Nassau County. The average size of the farms surveyed was 65 acres, of which 37 per cent- was in potatoes. On the average, 44 per cent of the total crop acreage was in potatoes, while the average potato acreage per farm was 24.8. The potato crop is relatively more important in the farming system here than elsewhere among the regions surveyed. The number of records taken on Long Island was 330, representing a total of 8188.16 acres planted to potatoes in 1912. The average yield per acre, on the farms surveyed, was 175.5 bushels. STEUBEN COUNTY The area of most intensive production in Steuben County lies in its northeastern part, along the Cohocton River valley and in the hill sections on each side. The Delaware, Lackawanna & Western and the Erie Rail- road handle the potato shipments. Local buyers take most of the crop from the grower, buying it either, at harvest time or on contract. They store it in temporary warehouses along the railroads or ship it direct. Because of the unevenness of topography and the heavy nature of the soil in this county, the highways are often so poor that the movement of the crop from field or cellar to the shipping point is seriously handicapped. For this reason, most of the crop is moved at the times when the roads are in the best condition. Much of it is shipped to New York and Philadelphia, but the variety Spalding's Rose 4 is sent to Florida as seed. The elevation of the surveyed fields ranged from 1200 to 2100 feet, the average beingv!659.2 feet. This wide range in elevation has considerable influence on the development of the potato crop, as is indicated by this study. A large part of the total crop is produced on hillsides of varying slope, the incline often being so steep as to limit the use of heavy machinery; on the other hand, many of the best potato fields are found on the level table-lands at the highest elevations. Northern Steuben County has • an average growing-season rainfall of from 16 to 18 inches, which is sufficient for maximum crops. Because of the heavy nature of the soil, years of abnormally large rainfall often cause much loss from blight rot. The growing season between killing frosts averages 150 days, and is usually sufficient to mature the crop. Because of better air drainage and cooler average temperatures, the crop is often later and the yields are larger on the farms at the higher elevations. This was not the case in 1912, however, as is shown later in the discussion of the influence of elevation. 1154 EARLE V. HARDENBURG Five soil series are principally concerned in the area studied in Steuben County, as shown by figure 129. Nearly half of the crop of 1912 was grown on Lordstown silt loam, which gave a higher average yield than any other series. In elevation this soil series is next to the Volusia series, which is found only at the highest elevations. Tho both of these soil series are naturally low in fertility, the highest average yield was obtained on the Lordstown series, while the lowest average yield was produced on the Volusia series. The soils on the hilltops are largely derived from shale and sandstone; the valley soils, altho naturally higher in fertility, contain more stone and gravel. Relatively long and fixed rotations are used in Steuben County, the com- monest being potatoes, oats, hay two years. Frequently the sod is left until long past its profitable stage for hay, with the result that the humus content remaining for the potato crop to follow is seriously depleted. Farms on which the sod was left down for the shortest period of years showed the highest yield, and vice versa. Sometimes wheat followed oats in the rotation, giving two successive years of grain. The wheat was used as the nurse crop. These farms showed a higher average yield of potatoes than did the farms using only one year of grain. This may have been due to the additional residual fertilizer left from the second year of grain, or possibly to production on better soil than is ordinarily devoted to potatoes. Buckwheat, in which Steuben County ranks second according to the United States census of 1909, is commonly used to follow old sod land that is being broken for potatoes or to break virgin land recently cleared. On the smaller potato farms, corn for grain or silage is grown in the rotation with potatoes. Land values are as low in Steuben County as anywhere in New York, for much of the land is infertile and rough, and little of it has been sold or rented in recent years. The estimated values ranged from $25 to $80 and more an acre, the average being about $50. The average size of the farms surveyed was 145.8 acres, 10 per cent of this being in potatoes. The per cent of total crop acres per farm in potatoes was 18'. The average yield per acre on the 360 farms surveyed, which represented a total of 5301.1 acres of potatoes, was 136.4 bushels. MONROE COUNTY ^ The potato section of Monroe County covers most of the region east, west, and south of Rochester. Potatoes are an important crop on most of the farms south of the fruit belt that extends across the northern border of the county abutting on Lake Ontario. Excellent railroad facilities pro- vide transportation for the marketing of the crop, loading stations being located on the New York Central, the Lehigh Valley, the Delaware, Lackawanna & Western, the Erie, and the Buffalo, Rochester, & Pittsburg Railroad. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1155 Elevation is not an influential factor in this region, since its variation is only between 400 and 1000 feet, the average being 592.5 feet. In general the topography is gently rolling, and in only a very few places is it suffi- ciently uneven to affect production or the usual cultural practices. The growing-season rainfall is somewhat less than that of the other regions, ranging normally from 14 to 16 inches. However, it is seldom insufficient for maximum production. Due to the tempering influence of Lake Ontario, the average growing season is 165 days, which is somewhat longer than that of the other regions except Long Island. The soils on which the potato crop is produced are principally of the Dunkirk and Ontario series, as shown in figure 129. Altho both of these soils are naturally fairly fertile, a study of comparative yields shows that, other things being equal, the Dunkirk soils gave the higher production. The soil map of Monroe County shows an especially wide range in soil types within each of these series. The cropping system of this region is usually a four-years rotation of potatoes and grain or another crop, oats, wheat, hay. Corn is most commonly chosen as the additional cultivated crop to be raised with potatoes, tho beans and cabbage are sometimes used. The value of potato land ranged from $50 to $250 an acre, the average acre value being $150. The farms surveyed averaged 112.07 acres in size, 11 per cent of the total acreage being in potatoes; and the importance of the crop is emphasized by the fact that 15 per cent of the crop acreage is in potatoes. The average yield per acre of the 1913 crop, for the 3728.25 acres of potatoes on the 300 farms, was 126.2 bushels. FRANKLIN AND CLINTON COUNTIES The areas of production in Franklin and Clinton Counties are two: one consists of a broad, level stretch of fertile valley land along the St. Law- rence River, extending across the northern end of Franklin County and over into Clinton County; the other consists of hill and valley farms on each side of the Saranac River, in central Clinton County. In both these areas the potato lands extend back into the foothills of the Adirondack Mountains. Most of the production centers in Franklin County are located along the Rutland Railroad, while the Delaware and Hudson Railroad handles most of the crop of Clinton' County. The greater part of the sur- plus is marketed in the eastern seaboard markets after the early crops of Long Island, New Jersey, and the South have been sold. A thriving trade in seed potatoes has been developed with Long Island, New Jersey, and southern points. Being in close proximity to the Adirondack Mountains, this region has a wide range in elevation. It varies from 300 to 1850 feet, the average for the farms surveyed being 1038.2 feet. The excellent yields obtained at the higher altitudes are due largely to the cool climate there afforded. 1156 EAKLE V. HARDENBURG In spite of the range in elevation, very little of the crop is produced on anything but level land. The farms along the St. Lawrence River valley are generally level or gently sloping toward the river, and most of the crop in Clinton County is also grown on fairly level fields, either in the Saranac River valley or on top of the foothills of the Adirondack^. Due to the northerly latitude of this region the growing season is rela- tively short, the average period between killing frosts being 150 days. Elevation and latitude are jointly contributing factors for an ideal potato climate conducive to late maturity of the crop. As a rule the growth is stopped by frost, resulting in a crop more or less immature at harvest time. This gives a product of excellent seed value and keeping qualities. The growing-season rainfall averages from 14 to 18 inches, the mountain areas receiving the greater precipitation. The rainfall is uniform thruout the growing season, each month averaging 3 or more inches. Most of the soils of this region are a fine sandy loam and are included in the Ontario, Calqma, and Terrace soil series. The Ontario series com- prises the area along the St. Lawrence River, and the Caloma and Terrace soils comprise most of the area in central Clinton County (fig. 129). The Ontario series is largely of sedimentary origin and its fertility is rather higher than the average; while the Caloma and Terrace soils are mainly of glacial drift formation and are of only mediocre fertility. The commonest system of cropping is a five-years rotation of potatoes and corn, oats, hay three years. The corn is used mainly for silage. Hops have been regarded as a relatively important cultivated crop in the Franklin County area until recently, when low prices, disease, and com- petition with the western crop caused a decided decrease in acreage. At present, potatoes are the chief source of cash income in this district. Land values here are similar to those in Steuben County, the range being from $10 to $100 an acre, with the average at about $50. The average size of the farms surveyed was 169.5 acres. Only 4 per cent of the total acreage, and 10 per cent of the crop acreage, was in potatoes. The average yield per acre on the 300 farms surveyed, repre- senting 2160 acres, was 179.3 bushels. METHOD OF STUDYING SURVEY DATA As previously pointed out, one of the handicaps in any effort to de- termine, by an analysis of survey data, the absolute influence of a single factor on yield, lies in the difficulty of separating the influence of other factors from that of the one in question. This is a necessary step, how- ever, in insuring accuracy and a correct interpretation of results. A preliminary study of factors influencing potato yield in Steuben County in 1912 (Hardenburg, 1915 b) indicated that the most important factors were the amount of seed used per acre, the value of manure and fertilizer employed per acre, and the frequency of bordeaux spraying. The results A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1157 of the present study have borne out that conclusion. Therefore, in con- sidering the influence of a given factor on yield, an effort has been made to eliminate as far as possible, or at least to give due credit to, other con- tributing factors. Since the study of each region concerns but a single year, too definite conclusions must not be drawn in interpreting the data presented. De- pending upon the normality of the season in which the crop was grown, the degree of influence of a given factor may or may not be maintained under average conditions. Tho cultural practices are not usually varied radically from one year to another, differences in the length of the grow- ing season, in the average growing-season temperature, and in rainfall, tend to affect the influence of those practices. Therefore it will not be possible to answer, in any appreciable degree, many of the questions that will be raised. The consideration of experimental results is there- fore of value in furnishing background for the study of each factor. As stated by Warren (1914), there are questions that can be answered only by a study of the results obtained on farms, and other questions that can be answered only by the results of experiments. Little attempt has been made to discuss any potato literature except that pertaining to seed, fertilizers, and planting, these being obviously the most influential factors under the grower's control. In reviewing the literature, one is impressed by the large quantity available and by the meagerness and unreliability of the data given to substantiate the state- ments. THE STUDY OF FACTORS CLIMATE A brief review of climatic conditions in each of the surveyed areas has been given, not because of any definite influence on the crop under con- sideration, but to make clearer the normal conditions to which the crop is subject. Facilities for taking weather data in each of these regions are not yet sufficient to allow of any attempt at the correlation of rainfall and temperature with yield for a given year. In general, the average growing-season temperature to which the crop is subject has a marked influence on- the vitality of that crop as used for seed. Briefly, high temperatures tend to produce devitalization. Long Island growers obtain average yields ranging from 150 to 250 bushels per acre from new Maine seed, but the use of the same stock for seed a second year results in greatly inferior yields, as is indicated in figure 131. The same principle is demonstrated in the rather common practice of introduc- ing seed from northerly latitudes, a practice which is justified on the basis of better yields, as is shown in the tests cited under the caption Source of seed. 1158 EARLE V. HARDENBURG FlQ. 131. GROWTH VARIATION BETWEEN NEW AND ONE-YEAR-OLD MAINE SEED STOCK ON LONG ISLAND The photograph shows also the characteristic topography of potato lands in Suffolk County Valuable studies of the influence of weather on the yield of potatoes in Ohio for a period of fifty-five years have been made by Smith (1915), and a similar study for a period of twenty-six years has been made in New York by Fox (1916). The relationship of both growing-season rainfall and temperature, in both States, is expressed in terms of the coefficient of correlation (r). A comparison of these coefficients shows that July is by far the most critical month with respect to these factors, in both Ohio and New York. The coefficient of correlation between tem- perature and yield is in most cases negative for both States, indicating that yield is inversely proportional to increase in temperature. So far as rainfall is concerned, the correlation for Ohio is positive and fairly large, indicating that rainfall is ordinarily a limiting factor in yield. The correlation of rainfall and yield in New York, on the other hand, is negative, showing that years of high rainfall are years of low yield. The average growing-season rainfall for the potato sections of New York, previously given as ranging from 14 to 20 inches, is evidently sufficient for this crop. The negative coefficient of correlation is probably a reflection of the fact that years of highest rainfall in New York have been years of severe loss from blight rot. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1159 ELEVATION Elevation as a factor influencing production has been determined from the figures shown on the topographic sheets of the surveyed areas pub- lished by the United States Geological Survey. As far as possible, the location of the potato fields for which data were taken was indicated on these topographic sheets at the time of taking the data. The chief difficulty in determining the absolute influence of elevation lies in the fact that increase or decrease in elevation is usually accompanied by a difference in soil type. A study of elevation, therefore, really involved also the consideration of both climate and soil. The writer is not aware that any test has ever been made in which either one or the other of these factors was studied with the other factor eliminated. Progressive increases in altitude and in latitude are similar in that each is accompanied by a reduction in temperature. The United States Weather Bureau, in computing temperature equivalents, makes use of the principle that every 300 feet rise in altitude is accompanied by a reduc- tion in temperature of one Fahrenheit degree. Influence of elevation on Long Island Elevation cannot be considered a potent factor in the Long Island area, for its highest point does not greatly exceed 200 feet. Many farms along the south shore of Suffolk County are below sea level, the sand dunes alone keeping out the sea. A typical Long Island potato field is shown in figure 131. The relation of elevation to yield in 1912 is shown in table 2: TABLE 2. RELATION OF ELEVATION TO YIELD ON 327 LONG ISLAND FARMS IN 1912 Elevation (feet) Number of farms Average yield per acre (bushels) i Average amount of seed used per acre (bushels) Average value of manure and fertilizer per acre Average elevation (feet) 1-50 157 178.3 12.9 $35.31 26.4 50 - 100 87 157.5 12.0 27.71 64.8 100-150 53 184.3 12.4 32.20 117.2 150 - 200 22 188.8 12.2 33.01 167.7 200 and over 8 196.3 13.4 31.65 218.1 Total 327 . Average . . . 175.3 12.5 $32.39 65.5 1160 EARLE V. HARDENBURG Altho there is no proof in table 2 that the yield increases with an increase in elevation, there is a slight indication that this may be true. The farms located at 50 to 100 feet elevation had a lower yield than those at the lowest elevation, partly because they received less seed and fertilizer than any other group. Furthermore, the farms at the lowest elevation received slightly more than the average amount of seed and fertilizer per acre. It is improbable, however, that the wide difference in yield between the two groups at the lowest elevations was due entirely to differences in amount of seed and fertilizer. There may have been some basic reason why the 87 growers at the 50-to-100-feet elevation used the least seed and the least fertilizer, which would account in part for the lower yield. No such reason is apparent, however, from the data at hand. Influence of elevation in Steuben County The average elevation of the farms visited in Steuben County is greater than in any other of the regions concerned in this survey, it being 1659.2 feet. The elevation varies from 1200 to 2100 feet, a range of 900 feet, and within this range there is a considerable variation in the soil types, as is shown later in table 13 (page 1770). A summary of the average yields obtained at various elevations is given in table 3 : TABLE 3. RELATION OF ELEVATION TO YIELD ON 355 STEUBEN COUNTY FARMS IN 1912 Elevation (feet) Number of farms Average yield per acre (bushels) Average unharvested yield per acre (bushels) Average elevation (feet) 1200-1300. . 9 148.8 24.6 1,243.6 1300-1400 36 156.6 13.1 1,336.0 1400 - 1500 39 129 7 20 5 1,426 2 1500-1600 46 136 4 20 0 1,530 4 1600 - 1700 46 133 9 24 1 1 , 630 . 2 1700-1800 . 63 131 6 29 0 1,732.1 1800-1900 . 61 138 3 30 7 1,829.7 1900 - 2000 .... 34 134 6 27.9 1,920.4 2000-2100 21 124 7 17.0 2,033.4 Total 355 Average 136 4 24 2 1 , 659 . 2 A general tendency for yields to decrease as elevation increases is ndicated by table 3. This is counter to the expected influence of altitude, and may be explained by the fact that the soil at the higher altitudes of A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1161 this region is heavier and of lower natural fertility. Further evidence of this condition is found in the figures showing a greater percentage of unharvested yield due to blight rot, which is so common in these heavier soils, at the higher elevations. The Green Mountain, or white-sprout, type of potato withstands less heat than does the Rural, or blue-sprout, type. Where the growing- season temperature is relatively cool, as in Franklin and Clinton Counties and on Long Island, the white-sprout type is therefore more common. In Steuben County, of 94 farms growing the white-sprout potatoes, 61 per cent were located above 1660 feet elevation and only 39 per cent were located below this level. Of 239 farms growing the blue-sprout type, 50 per cent were above and 50 per cent were below 1660 feet elevation. There is some tendency, therefore, to grow more of the white-sprout type at the cooler altitudes. Influence of elevation in Monroe County The range of elevation in Monroe County is between 400 and slightly over 800 feet. Little opportunity is therefore afforded to study the influence of this factor in this region. The figures in table 4, interpreted in the light of average seed and fertilizer used, show that elevation has some tendency to increase yield. TABLE 4. RELATION OF ELEVATION TO YIELD ON 296 MONROE COUNTY FARMS IN 1913 Elevation (feet) Number of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer 400-500 . 30 130 5 12 6 $14 03 500-600 . 107 122 8- 12 5 11 20 600-700. . 129 116 0 12 4 10 86 700 - 800 .. 23 •165 2 12 5 11 56 800 and over 7 225 4 14 7 11 39 Total .... 296 Average . . 127 1 12 5 $11 34 Influence of elevation in Franklin and Clinton Counties A variation of over 1500 feet elevation in the farms in Franklin and Clinton Counties affords excellent opportunity for the study of the influence of elevation on yield. The summary given in table 5 shows a rather marked 1162 EARLE V. HARDENBURG TABLE 5. RELATION OF ELEVATION TO YIELD ON 290 FKANKLIN AND CLINTON COUNTY FARMS IN 1913 Elevation . (feet) Number of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer Average elevation (feet) 300- 600 33 154 8 11 6 $12 90 437 7 600-800 19 154 9 10 9 11 12 697 1 800 - 1000 . 31 185 1 11 5 13 09 895 5 1000 - 1200 ... 47 184 0 12 1 13 10 1 094 4 1200-1400 101 179 1 12 2 12 72 1,296 9 1400-1600 42 191 0 12 5 13 22 1,468 6 1600-1850 17 193 4 11 8 14 13 1,709 7 Total 290 Average 177 3 11 9 $12 91 1 038 2 influence of this factor. With the amounts of seed and the value of manure and fertilizer used approximately equal, the best yields were produced at the higher mountain elevations. Since the increase in elevation for this 'region is accompanied by a con- siderable variation in soil type, a part of the increase in yield at the higher levels may be due to the latter factor. However, since Franklin County has not been soil-surveyed, it is impossible here to measure accurately the influence of the soil. Very little difference in soil type was evident between the Dover fine sandy loam of the lower elevations and the Caloma fine sandy loam of the higher elevations in Clinton County. CROP ROTATION The benefits of crop rotation to a heavy-feeding cultivated crop such as potatoes have long been recognized. The crop survey as a means of com- paring various rotations in a given region, however, has very limited possi- bilities, for in the older farming regions the same general type of rotation is followed thruout. Very few tests have thus far been made by the experi- ment stations to determine the most suitable place in the rotation and the best length of rotation for potatoes in a given region. Probably the most valuable work has been done by Hartwell and Damon (1916) in their twenty-years comparison of different rotations of corn, potatoes, rye, and grass, at the Rhode Island Station. The principal feature of this work lies in 'a comparison of four-, five-, and six-years rotations of potatoes, jye and rowen, grass, corn, the grass being left down for from one to three years. No stable manure was used, but complete commercial fertilizers A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1163 were added to the sod each year. In the matter of fertilizers, Hartwell and Damon's experiment is not comparable to farm practice in New York, where little or no commercial fertilizer is ever used, stable manure being generally applied, instead, as a top dressing, during the last year of sod or perhaps just before plowing for corn or potatoes. The average yields per acre of potatoes obtained by Hartwell and Damon, in the rotations including grass for one, two, and three years, were 200, 199, and 223 bushels, respectively. It appears that their commercial-fertilizer treat- ments were sufficient to maintain a maximum condition of sod thruout the three years. A test on the influence of various fertilizers on potatoes, conducted at the Rothamsted station, is reported by Hall (1905). In this test the crop was grown for twenty-six consecutive years on the same land, and under each treatment the yields declined during the later as compared to the earlier years of the test. Long Island is the only section in New York in which the crop is grown without rotation, and it is only the in- creased use of fertilizers that has maintained yields there. Not only is it difficult to get sufficient stable manure for the potato crop on Long Island, but many growers do not find it economical to haul fertilizer in this form so great a distance as would often be necessary. Consequently, each year more than a third of the growers sow a cover crop of rye after potatoes. Some use the cover crop every year, while others use it only every second or third year, and some not at all. In the consideration of the influence of cover crops on yield, only those fields are included on which a cover crop was grown in the fall and winter preceding the potato crop. In table 6 the average yields that are obtained directly after cover crops, are com- pared with those obtained when no previous cover crop had been used. TABLE 6. RELATION OF COVER CROP TO YIELD ON 313 LONG ISLAND FARMS IN 1912 Treatment Number of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer per acre Cover crop ' 131 174 1 12 6 $32 61 No cover crop 182 177 3 12 5 32 25 Total . 313 Average 175 5 12 5 $32 40 The figures given in table 6 should not be construed to mean that cover crops are not beneficial to the potato crop on Long Island, because the 1164 EAELE V. HARDENBURG yields obtained in the group listed as not using a cover crop may have been produced on farms which used a cover crop two or three years previously or on farms whose soil was naturally higher in organic content. Granting vthis, the data on cover crops for Long Island are not sufficient to indicate either advantage or disadvantage accruing from its use. It is true that in 1912 growers who had not sown a cover crop the previous fall did not attempt to supplement the soil fertility by using more fertilizer. This in itself may indicate that, in the main, only those growers who actually needed the cover crop to maintain yields were the ones who used it. The rotations followed in Steuben County, consisting usually of potatoes, grain, and hay, vary principally in the number of successive years that the hay and the grain are left on the same ground. Commercial fertilizer is applied lightly at the time of planting potatoes, and, altho what stable manure is available is put on the sod to be plowed for potatoes, there is seldom enough to cover the entire potato acreage. The yields of hay are largely dependent on the residual fertilizer left from that applied directly to the grain crops. Thus in the longer rotations, in which sod is left down for three or more years, only a poor supply of root and stubble residue is left to supply humus to the potato crop. A comparison of the influence on the yield of various types of rotations in this region is shown in table 7: TABLE 7. RELATION OF ROTATION TO YIELD ON 240 STEUBEN COUNTY FARMS IN 1912 Manure or fertilizer on Manure or fertilizer on No manure nor fertilizer part of acreage entire acreage used Rotation Num- ber of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Num- ber of farms Average yield per acrej (bushels) Average cost of manure and ferti- lizer Num- ber of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Potatoes, grain, hay. . 13 177.0 10.9 8 189.1 $10.51 1 150.0 7.5 Potatoes, grain, hay, hay 117 134.9 10.3 58 150.1 13.98 7 106.2 9.6 Potatoes, grain, hay, hay, hay 62 122.7 10.0 26 135.0 11.95 3 103.6 10.5 Potatoes, grain, grain, hay, hay 25 150.1 9.2 12 160.9 13.34 Potatoes, grain, grain, hay, hay, hay 11 143.0 8.8 5 160.2 20.34 1 50.0 8.8 Eliminating the factors of seed and fertilizer as given in table 7, the yield consistently decreased with each successive year that the sod remained in rotation. This shows the tendency of the seeding to become thinner and of less value as a source of humus for the potato crop, the older it becomes. The figures for the last two rotations in the table — which differ from the first three in that they contain two years of grain instead of one, and from A STUDY OF FACTORS INFLUENCING, THE YIELD OF POTATOES 1165 each other only in the number of years of successive hay crops — show a higher average yield of potatoes with them than with the first three. This may be due to the additional residual fertilizer left from that applied to the extra year of grain, or to the factor of naturally better soil as indicated by the tendency to produce more grain. The type of rotation commonest in each region is indicated by the figures in table 8 on the percentage of total crop acres occupied by each crop listed. No fixed rotation is indicated, for Long Island, where potatoes are grown for a varying number of successive years on the same land. The TABLE 8. RELATIVE IMPORTANCE OF CROPS ON FARMS SURVEYED Crop Per cent of crop acres Long Island Steuben County Monroe County Franklin and Clinton Counties . Average Hay Potatoes Oats Corn for grain Wheat ,. Orchard Rye Corn for silage Cabbage Sweet corn Beans 12 43 2 15 5 0 1 1 4 5 1 0 5 0 4 2 0 0 0 0 0 42 18 21 1 3 1 4 0 0 0 0 4 0 2 0 0 1 1 1 1 0 24 15 17 6 14 8 4 2 3 1 4 0 0 1 0 0 0 1 0 0 0 59 10 19 1 0 2 1 4 0 0 0 1 0 1 0 0 1 0 0 0 1 34.25 21.50 14.75 5.75 5.50 2.75 2.50 1.75 1.75 1.50 .25 .25 .25 .00 .00 0.50 0.50 0.50 0.25 0.25 0.25 Buckwheat Cauliflower Barley Garden truck Brussels sprouts Corn for fodder Alfalfa Peas Oats and barley Sugar bush • figures for Steuben County indicate a rotation of potatoes, oats, hay two years; those for, Monroe County, a rotation of potatoes with corn or beans or cabbage, oats, wheat, hay one to two years ; and those for Franklin and Clinton Counties, a rotation of potatoes with corn, oats, hay three years. A review of the experimental literature on the influence of crop rotation in potato production shows a striking preference for either grass, or a 1166 EARLE V. HARDENBURG legume productive of considerable vegetative growth, as a crop to precede potatoes. This is evidence of the efficient use which the potato crop is able to make of this form of organic material. Such legumes as cowpeas, soybeans, and crimson clover commonly precede potatoes in the Southern and the South Atlantic States, while timothy, in combination with red or alsike clover, is used generally thruout the principal potato States. Alfalfa is considered the ideal legume to precede potatoes in the alfalfa belt of the West. The root and stubble residue from these crops not only contributes to the food requirements of the potato, but also improves the aeration, the temperature, and the moisture-holding ability of the soil. Generally speaking, the rotations of the three regions aside from Long Island are long enough not to serve as factors limiting yield except as the type of rotation may affect fertilizing practices. Inasmuch as the avail- able stable manure is not usually applied for the benefit of the hay crops, and the residual organic fertility is not thereby maintained or improved, the sod residue commonly turned under before potato planting is usually less Valuable after a three-years stand than after a stand of shorter duration. VALUE OF LAND The farmer's estimate of farm land values is very often not based on productive value, altho this factor, together with the distance from rail- road and city and the salability of the farm, usually enters into the appraise- ment. A correlation of estimated value with average yields will show, in a measure, the extent to which productive ability of potato land enters into its evaluation. App (1916), studying the factors that influence farm profits on potato farms in Monmouth County, New Jersey, found a consistent tendency for farm acre values to decrease as distance from the railroad increased. His similar conclusions with respect to crop acre values and labor income, however, do not seem warranted from the data given. The figures obtained on land values in the regions surveyed represent the estimated selling value of potato land only. A more important factor than the distance from the post office, which was ascertained and used in making this estimate, would have been that of the distance from the nearest city, village, or railroad. Apparently, on Long Island, land valued up to $550 an acre is yielding an increased crop with the increase in value (table 9). However, it is true also that the increase in land values is accompanied by the use of more seed and more fertilizer, and by more spraying for blight. These combined factors would easily account for the consistent increase in yield. The farms showing a land value of over $550 an acre are located prin- cipally in Nassau County, at a considerable distance from the post office, and are appraised at their real-estate value. In fact, much of the land has been sold at fabulous prices for real-estate purposes and is now rented A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1167 TABLE 9. RELATION OF VALUE OF LAND TO YIELD ON 330 LONG ISLAND FARMS IN 1912 Value Num- ber of farms Average yield per acre (bushels) Average distance from post office (miles) Amount of seed used per acre (bushels) Value of manure and fertilizer per acre Per cent of farms using bordeaux Less than $250 $250-8400 $400-$550 118 132 27 147.0 184.7 196.7 2.7 2.3 2.8 12.1 12.6 13.0 $30.24 33.50 35.24 24 45 52 $550 and over '. . . . 53 191.9 5.6 12.7 32.73 11 Total 330 Average 175 5 3 0 12 5 $32 40 32 back to the original owner who is again growing potatoes on it. Some- what less seed and fertilizer are used on these farms, and less spraying is done on them. In Steuben County, potato land valued up to $80 an acre gives increased yields with the increase in value (table 10). Tho the amount of seed used TABLE 10. RELATION OF VALUS OF LAND TO YIELD ON 360 STEUBEN COUNTY FARMS IN 1912 Value Num- ber of farms Average yield per acre (bushels) Average distance from post office (miles) Amount of seed used per acre (bushels) Value of manure and fertilizer per acre Per cent of farms using bordeaux $25-$40 46 110 2 5 3 ' 9 6 $ 7 93 4 $40-$50 111 134 1 4 2 JO 4 9 01 6 $50-$60 76 139 4 3 6 10 1 10 86 8 $60-$70 44 144 4 3 6 10 4 11 80 o $70-$80 42 148 2 2 9 10 3 12 29 o $80 and over 41 145 1 2 5 9 5 10 31 5 Total . . 360 Average . 136 4 3 8 10 1 $10 14 5 per acre is about the same thruout, there is a tendency to spend more in manure and fertilizer for the higher-priced land. The real reason for 1168 EARLE V. HARDENBURG this is doubtless the relative cheapness with which manure and fertilizer can be handled by the growers nearest the villages. The increased yield of the higher-priced land may be due in part to this increase in the value of manure and fertilizer used. Land values decrease as the distance from the post office increases, in Steuben County. Distance, in fact, may largely determine the valuation of potato land. In Monroe County, as in Steuben County; the yields increased with the increase in land values (table 11), the yield increase being accompanied by, TABLE 11. RELATION OF VALUE OP LAND TO YIELD ON 297 MONROE COUNTY FARMS IN 1913 Value Num- ber of farms Average yield per acre (bushels) Average distance from post office (miles) Amount of seed used per acre (bushels) . Value of manure and fertilizer per acre Per cent of farms using bordeaux $ 50- $100. . 27 103 4 4 3 11 1 $10 86 19 $100- $150 V. 145 128 3 2 6 12 5 10 09 23 $150- $200 69 130 6 2 3 12 7 12 02 35 $200- $250 56 129.3 2.3 13.0 14.10 18 Total 297 Average 127 0 2 6 12 5 $11 33 24 TABLE 12. RELATION OP VALUE OP LAND TO YIELD ON 303 FRANKLIN AND CLINTON COUNTY FARMS IN 1913 Value Num- ber of farms Average yield per acre (bushels) Average distance from post office (miles) Amount of seed used per acre (bushels) Value of manure and fertilizer per acre Per cent of farms using bordeaux $ 10-$ 25.. 29 160 2 4 2 10 2 $12 35 0 $ 25-$ 50 105 173 4 3 5 11 3 13.73 0 $ 50 - $ 75 114 189 0 3 7 12 9 12 79 2 $ 75 - $100 . 42 178 2 2 5 12 3 11 83 0 $100 and over. 10 176 9 1 6 13 0 15 02 10 Total 300 Average 179 3 3 4 12 0 $13.01 1 A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1169 and doubtless largely due to, an increase in the amount of seed and in the value of manure and fertilizer used. Here also the land values tend to decrease as the distance from the post office increases. In Franklin and Clinton Counties there is a tendency to spend more for seed, fertilizer, and spraying, on the farms having the higher- valued potato land (table 12). This expenditure is apparently justified on land valued up to $75 an acre. Above that point, the average yield did not increase even with increased expenditure. This may be taken as an indication that such land was valued at more than its productive ability would justify. These more valuable farms are situated near Peru, in Clinton County, and are thus highly valued because of their location in the apple section of northern New York rather than on the basis of their adaptability to potato culture. SOIL It has not been possible to study the influence of soil on yield on Long Island and in Franklin and Clinton Counties. Altho Clinton County has been soil-surveyed, it is included in the tabulations with Franklin County, and the records taken were insufficient to justify the making of such a study on Clinton County alone. The Monroe County soils map, published by the United States Bureau of Soils, has been used in correlat- ing yield and other factors with the soils of that region. Professor E. 0. Fippin, formerly of the Department of Soil Technology at Cornell University, accompanied by the writer, made a reconnaissance soil survey of the surveyed area in Steuben County in the summer of 1916. Professor Fippin's familiarity with soil mapping in New York enabled him therefor to sketch the boundaries of the various soil types and series on the topographic sheets previously used in locating the surveyed potato fields. Nearly half of the crop in Steuben County is growri on the Lordstown soil series at an average elevation of 1718.2 feet. As indicated in table 13, the highest average yield was obtained on this soil series in spite of the fact that only an average amount of seed was used and somewhat less than the average value of manure and fertilizer. In contrast to this, the Volusia soil series, located on the hilltops at an average elevation of 1785.5 feet, yielded the lowest average yield of any series in spite of the fact that about the average amount of seed was used and more than the average value of manure and fertilizer. The principal difference between these two soil series lies in the somewhat darker color and the better ox'dized condition of the Lordstown series. It is true that the soils of this region become lighter in texture and of higher gravel and stone content as the valleys are approached. This condition is probably blamable, at least in part, for the higher percentage of blight rot on the heavier soils 1170 EARLE V. HARDENBURG at the higher elevations, as indicated in table 13. Altho very little spray- ing for blight was done in 1912, it was noted that the fields which were sprayed returned a higher average yield than those which were not sprayed, irrespective of the soil type. Apparently, depth of planting and date of planting are not influenced by soil type. TABLE 13. RELATION OF SOIL TYPE TO YIELD ON 293 STEUBEN COUNTY FARMS IN 1912 Soil type Num- ber of farms Average yield per acre (bushels) Average unhar- vested yield per acre (bushels) Average value of manure and ferti- lizer Average amount of seed used per acre (bushels) Per cent of farms using bordeaux Average depth of planting (inches) Aver- age date of plant- ing Average elevation (feet) Volusia silt loam and loam Lordstown silt loam Wooster gravelly loam Rodman gravelly loam Chenango gravelly loam 36 151 19 59 28 115.9 144.8 126.9 142.1 140.4 32.7 25.2 16.0 24.0 23.7 $10.79 9.35 11.22 11.76 9.25 10.2 10.9 9.2 10.1 11.0 0 6 0 2 7 3.4 3.1 3.2 3.0 3.1 May 17 May 19 May 17 May 22 May 18 1,785.5 1,718.2 1,637.9 1,496.5 1,364.6 Total 293 Average 139 6 25 1 $10 08 10 6 4 3 1 May 19 1 642 8 In the area surveyed in Monroe County, four soil series are concerned — Ontario, Dunkirk, Clyde, and Genesee. In all, fourteen soil types are involved, but because of the small number of farms on some of these types, only those shown in table 14 are used in correlating soil with yield and other factors. On the basis of seed used, of value of manure and ferti- lizer, and of percentage of farms using bordeaux, the Dunkirk fine sandy loam and the Dunkirk fine sand are naturally the best for potatoes from the standpoint of yield, among the types considered. Altho in 1913 nearly half of the crop in the surveyed area was grown on Ontario fine sandy loam, under at least average cultural treatment, it gave the lowest average yield per acre of any series studied. As indicated in the summary of table 14, the soil types of the Dunkirk series seem to give higher yields than those of the Ontario series. The average amount of seed, fertilizer, and spraying was about the same in both series. Here, as in Steuben County, depth and date of planting do not seem to be influenced by any difference in soil type. The average elevation of the two soil series sum- marized is almost identical. In brief, with more seed, more fertilizer, and a higher percentage of area sprayed with bordeaux, the Ontario soils yielded 20 bushels per acre less than did the Dunkirk soils. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1171 TABLE 14. RELATION OF SOIL TYPE TO YIELD ON 253 MONROE COUNTY FARMS IN 1913 Average Average Soil type Num- ber of Average yield per value of manure amount of seed used Per cent of farms Average depth of Average date Average elevation farms acre (bushels) and ferti- per acre using bordeaux planting (inches) of planting (feet) lizer (bushels) Dunkirk fine sandy loam 20 186 6 $11 45 12 4 15 2.9 June 14 659.3 Ontario fine sandy loam 124 .122.0 11.09 12.6 19 2.9 June 3 595.2 Dunkirk gravelly loam 32 123.1 9.83 12.0 25 3.1 June 7 586.2 Ontario loam 52 124.2 12.99 13.0 6 3.3 June 6 582.0 Dunkirk gravelly sandy loam 11 123.7 9.01 11.0 27 3.3 June 8 573.6 Dunkirk fine sand. . . 14 137.2 11.24 12.4 21 3.7 June 7 515.0 Total 253 Average 128 3 $11 31 12 5 17 3 1 June 6 591 0 Summary Ontario series 176 122.7 $11.65 12.7 27 3.0 June 5 591.3 Dunkirk series 77 142.6 $10.42 12.0 17 3.2 June 10 590.5 PLOWING The study of plowing aS to its bearing on yield is treated under two head- ings — time of plowing and depth of plowing. The average date of plow- ing was obtained by averaging the actual dates of plowing for each region. It is recognized, of course, that this date will vary from year to year, depending on seasonal conditions. The only region in the State in which fall plowing was done to any appreciable extent was Franklin and Clinton Counties. To determine the depth of plowing, the grower was in each case asked to give his estimate of the actual depth, in inches, which he plowed for potatoes. Time of plowing For many years, such advantages as increased liberation of plant food, elimination of injurious insects, and improved soil texture, have been pointed out in favor of fall plowing. Very little experimental evidence has been presented, however. Dickens (1914) has furnished data covering two years of work at five substations in Kansas, all showing a decided increase in yield on fall-plowed as compared to spring-plowed land for potatoes. A summary of the data on the season at which potato land was plowed in each of the surveyed regions in this study is given in table 15: 1172 EARLE V. HARDENBURG TABLE 15. PER CENT or POTATO LAND FALL- OR SPRING-PLOWED IN THE AREAS SURVEYED Franklin Long Steuben Monroe and Time of plowing Island, County, County, Clinton 1912 1912 1913 Counties, 1913 Fall 1 5 1 7 0 33 57.0 Part fall and part spring 2 1 8 0 0.00 16.0 Spring 96 4 90 3 99 67 27 0 The average dates of spring plowing in 1912 for Long Island and for Steuben County were April 3 and April 29, respectively. The average dates of spring plowing in 1913 for Monroe County and for Franklin and Clinton Counties were May 15 and May 12, respectively. No speculation as to the reason for the greater proportion of fall plowing in Franklin and Clinton Counties is offered, but it is presumed to be due as much to weather conditions for the year as to labor competition with work on other crops. The comparative yields on land plowed at the different seasons in Franklin and Clinton Counties are shown in table 16: TABLE 16. RELATION OF TIME or PLOWING TO YIELD ON 300 FRANKLIN AND CLINTON COUNTY FARMS IN 1913 Time of plowing Num- ber of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer Average acreage of potatoes Fall 171 183 7 12 4 $12 81 7 2 Part fall and part spring Spring 48 81 173.9 173 2 11.6 11 5 14.47 12 46 7.2 7 2 Total 300 Average . . . 179 3 12 0 $13 01 7 2 In view of the fact that about a bushel more of seed and a slightly higher value of manure and fertilizer were used on the fall-plowed land, the dif- ference in yield of 10.5 bushels per acre in favor of fall plowing may not be entirely due to a difference in the time of plowing. There is no indication that fall plowing is commoner on the larger potato acreages, since the average acreage was the same in all three groups. A STUDY OF FACTORS INFLUENCING THE YIELD* OF POTATOES 1173 Depth of plowing ' It might be expected that a crop such as the potato, which develops underground and is subject to varying conditions of soil moisture and soil texture, would be influenced by the factor of depth of plowing. How- ever, no actual experiments with this problem have come to the writer's attention. Dickens (1914) states that shallow plowing has given the best results on loamy soil at the Kansas station. Stone (1905) states that at the Cornell station, deep plowing on the Dunkirk gravelly soil gave the best results. Generally speaking, shallow plowing has been recommended for heavy soils and deep plowing for light soils. The possibility of drawing definite conclusions from a study of a factor depending so much on the grower's estimate and on only one year's results, is necessarily limited. This is one of the factors that for its ultimate solu- tion must depend upon carefully controlled experiments on a given soil type in each region concerned. From the following discussion of the regions herein considered, it would appear that depth of plowing is an important factor only on soils of either extreme of texture — deep planting increasing the area for tuber development in heavy soils, and providing for planting at the moisture-table depth in light soils. Depth of plowing on Long Island Altho the soils of Long Island are noticeably lighter than those of any other potato region in the State, a marked correlation of depth of plowing with yield is evident in table 17. An increase in the depth of plowing was accompanied by the use of more seed and fertilizer per acre and a greater TABLE 17. RELATION OF DEPTH OP PLOWING TO YIELD ON 328 LONG ISLAND FARMS IN 1912 Depth of plowing (inches) Num- ber of farms Average yield per acre (bushels) Average amount of seed used ' per acre (bushels) Average value of manure and fertilizer Average number of times sprayed with bordeaux Less than 5 26 163 7 11 4 $30 01 2 5 5-6.. 60 162 9 11 6 30 53 3 o 6-7.. 119 173 8 12 8 32 15 3 0 7-8.. 65 174 1 12 9 33 65 3 3 8-9 36 188 6 12 7 35 26 3 4 9 and over 22 202.1 12.9 32.13 3.6 Total 328 Average, 6.3 inches. 175 6 12 5 $32 42 3 1 1174 EARLE V. HARDENBURG frequency of spraying. These factors contributed in some degree to the greater yield apparently resulting from the deeper plowing. In answer to the question whether the deeper plowing was accompanied by deeper planting, it may be stated here that, whereas the Long Island fields were plowed at about the same average depth as those of the similarly light soils of Franklin and Clinton Counties, the average depth of planting was deepest on Long Island, and shallowest in Franklin and Clinton Counties, of the four regions. Apparently, deeper plowing on Long Island is to be advised. By virtue of its descriptive value as well as its value as a means for measuring correlation, the biometrical method has been applied to this factor of depth of plowing for Long Island, as also for the other regions, and the result is shown in figure 132. The correlation coefficient, r, here Yield per acre, in bushels OjpO*O O iO Q *O O *O O »O "jt^OCM lO b- O C^l ^O t^OtN .3 3 & 4 5 || « O O . H O o- 9 £ 10 & 11 Q 12 * - & i o § i £ i i i i s s 1 1 2 i 5 4 1 9 5 2 21 2 5 15 12 6 10 5 2 1 2 60 1 2 5 11 33 16 19 11 8 8 5 119 2 5 4 16 11 12 5 7 3 65 5 5 7 3 8 4 1 1 1 1 36 1 1 1 8 1 2 • 1 15 1 2 1 1 1 6 0 1 1 1 6 27 37 81 42 60 27 23 13 83 328 r = 0.159 ± 0.036 FlG. 132. CORRELATION OF DEPTH OF PLOWING AND YIELD ON 328 LONG ISLAND FARMS IN 1912 has the value 0.159 d= 0.036. Inasmuch as the coefficient is positive and is more than three times its probable error, it shows a significant degree of correlation between depth of plowing and yield. Present-day biometri- cians are now well agreed that the significance of a coefficient is measured not alone by its absolute value, but in the light of its consistency with coefficients of other series or other years and its probable error. Depth of plowing in Steuben County In contrast to soil conditions on Long Island, the soils of Steuben County are the heaviest of any of the four regions. Nevertheless, a rather marked positive correlation between depth of plowing and yield is shown in table 18. In this region, as on Long Island, the growers who plowed deeper A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1175 TABLE 18. RELATION OF DEPTH OF PLOWING TO YIELD ON 360 STEUBEN COUNTY FARMS IN 1912 Depth of plowing (inches) Number of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer 4- 6 . 52 125.8 9.9 $ 8 59 6- 8 ... 258 136.2 10.1 10 39 8-10 ...... 50 149.4 10.4 10 54 Total . 360 Average, 6.6 inches 136.4 10.1 $10.06 for potatoes were inclined to plant more seed and to use a greater value of manure and fertilizer. The difference in yield of nearly 24 bushels per acre resulting from a difference of 4 inches in depth of plowing is evidently due, in part at least, to the increase in depth of plowing. The coefficient of correlation shown in figure 133 is 0.190 ± 0.034, a value expressing Yield per acre, in bushels 8 g 24 2 1 2 2 7 |/S5 1 4 7 11 14 5 2 44 £.S6 2 13 31 32 23 13 12 4 3 133 J.S7 1 2 7 18 19 26 26 18 4 4 1 126 a bc8 1 2 5 10 12 6 6 1 3 1 47 Q'S9 1 1 1 3 2 5 26 64 73 78 52 38 9 10 0 2 0 0 1 360 r = 0.190 d= 0.034 FlG. 133. CORRELATION OF DEPTH OF. PLOWING AND YIELD ON 360 STEUBEN COUNTY FARMS IN 1912 significant correlation. Most of the crop in this region is grown on soils underlain at rather shallow depths with more or less impervious strata. Deeper plowing under these conditions would tend to enlarge the area adapted to maximum tuber development. Depth of plowing in Monroe County The soils of Monroe County may be considered intermediate in texture between those of Long Island and those of Steuben County. As a rule, they are deeper than those of the latter region. A study of table 19 shows no 1176 EARLE V. HARDENBURG TABLE 19. RELATION OF DEPTH OP PLOWING TO YIELD ON 261 MONROE COUNTY FARMS IN 1913 Depth of plowing (inches) Num- ber of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer 4-7 47 136 4 11 7 $11 95 7-8 93 134 2 12 2 10 97 8-9. . . 90 132 5 12 7 12 05 9 and over . 31 145 0 14 5 11 42 Total . . 261 Average, 6.6 inches. . . 135 6 12 6 $11 57 apparent relation between depth of plowing and yield. With approxi- mately the same fertilization for each depth of plowing, the slight tend- ency for increased yields at the deeper plowing may easily be attributed to the larger amount of seed planted. The coefficient of correlation shown in figure 134 is 0.006 ± 0.039. Both Yield per acre, in bushels vO O 8 1 1 16 30 11 20 13 10 3 1 M 9 Q'S 10 2 4 1 8 6 9 1 1 2 1 2 1 1 2 52 100 106 33 5 299 1 7 38 7(X 50 60 34 24 8 4 0 1 1 r = 0.006 d= 0.039 ' FlG. 134. CORRELATION OF DEPTH OF PLOWING AND YIELD ON 299 MONROE COUNTY FARMS IN 1913 the coefficient and its probable error indicate alack of relationship between depth of plowing and yield for this region. Apparently the minimum depth of plowing was sufficient for maximum production in Monroe County soils. Depth of plowing in Franklin and Clinton Counties The figures shown in table 20 indicate that depth of plowing does not influence yields in Franklin and Clinton Counties. Tho the amount of seed A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1177 TABLE 20. RELATION OF DEPTH OF PLOWING TO YIELD ON 299 FARMS IN FRANKLIN AND CLINTON COUNTIES IN 1913 \ Depth of plowing (inches) Num- ber of farms Average yield per acre (bushels) Average amount of seed used per acre (bushels) Average value of manure and fertilizer 4- 5 23 175 5 11 3 $14 51 5- 6 70 180 3 11 4 12 37 6- 7 118 179 2 12 3 13 33 7- 8 52 180 4 11 9 13.38 8-10 . 36 179 6 12.7 12 07 Total 299 Average, 6.2 inches 179 3 12 0 $13 01 planted was increased slightly as the depth of the plowing was increased, the amount of fertilizer used was not increased. Consequently there would be ample opportunity for any influence of depth of plowing to be reflected in the yields under this method of study. The coefficient shown in figure 135 is 0.028 db 0.039, and indicates no relationship between the depth of plowing and the yield. Yield per acre, in bushels III r!? 5 ,? c5 i-H I o Q i— i i— i i 3 • I 7 § 05 L >o CM I 1 5 2 2 1 1 1 3 13 5 4 3 11 23 8 4 1 6 16 17 8 5 8 19 32 10 6 1 6 16 10 5 2 3 8 8 4 1 1 5 5 1 2 2 2 1 26 14 23 70 118 52 32 ;4 299 12 26 50 52 76 38 . r = 0.028 ±0.039 FlG. 135. CORRELATION OF DEPTH OF PLOWING AND YIELD ON 299 FRANKLIN AND CLINTON COUNTY FARMS IN 1913 MANURE AND FERTILIZER The commercial-fertilizer industry, as a country- wide enterprise, began in the Eastern States and dates from about 1860. Previous to that time, the potato crop depended for its plant food largely upon the natural available supply of the soil, supplemented by applications of barnyard manure. Manure has been recommended by many experiment stations as perhaps the best source of nitrogenous plant food for this crop. How- ever, as the acreage increased in the East and the soils became more 1178 EARLE V. HARDENBURG impoverished, the need for a commercial source of plant food became imperative. Today there are few crops which require more and respond be'tter to fertilizer than do potatoes, tho even yet commercial fertilizer is used very little on the newer potato lands of Michigan, Wisconsin, and Minnesota. From the beginning of the fertilizer industry, hundreds of tests have been conducted by the eastern state experiment stations to determine the influence on the yield of potatoes of such factors as the amount of fertilizer used, its analysis, and the time and method of its application. According to Whitney (1910)v 1769 such tests were conducted between 1868 and 1908, a period of forty years. Of all the tests made up to 1908, nearly 72 per cent fall within the ten-years period from 1890 to 1900. Twenty-three States contributed to these tests, and about 57 per cent of the total were made in New York, Ohio, and New Jersey. Whitney states that it is impossible to draw conclusions even from an average of simi- lar experiments among those listed, since the variation in the yields of check plots of individual experiments sometimes ranges as high as 900 per cent. The crop survey has been found to have its limitations in the study of such questions as best analysis, best amount, or best source of ingredients, of a fertilizer to be used for potatoes. It is generally impossible to get information from the grower as to the analysis or the source of the elements of the fertilizer he has used. Many growers who were .questioned had been more impressed by the brand name or by the price paid for the fertilizer than by its analysis. An attempt to correlate the amount of fertilizer per acre with the yield resulting was found impracticable with- out knowledge of its analysis, owing to the fact that large applications of a cheap fertilizer might be no more than equivalent to small applications of a high-grade fertilizer. Furthermore, many growers used manure in place of fertilizer, or vice versa, while many others used both on the same acreage. The study of the influence of manure and fertilizer on yield in the surveyed regions has therefore been made on the basis of the combined value per acre of manure and fertilizer. Estimates of the value of the manure used, made by the grower, and the prices he paid for fertilizer, have been used. In determining the proportion of the total value of the manure received by the potato crop, depending on the time and place of its application, 50 per cent of its value was charged if it was applied directly to the potato crop, 30 per cent if it was applied to the crop just preceding the potatoes, and 20 per cent if it had been applied two years before potatoes. This evaluation of residual manure is not based on exact experimental knowledge, but is presumed to represent the approximate availability of stable manure for successive crops. The Department of Agricultural Economics and Farm Management at Cornell University estimates that, on the heavier soils, 40, 30, 20, and 10 per cent of the value of manure is received by the first, the second, the third, and the fourth crop after its application, respectively. On lighter soils, which A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1179 are more subject to leaching, probably a charge of 50 per cent to the first crop would be more nearly correct. A uniform basis of evaluation has been applied in this study. Home-mixed fertilizers Potato growers have never adopted the practice of mixing their own fertilizer, even in the regions where potatoes are raised on a very extensive scale. The advantage of using home-mixed fertilizers has usually been con- sidered, to lie in a saving in cost rather than in an increase in yield. Woods and Bartlett (1909) compared several home-mixed fertilizers with a common ready-mixed commercial fertilizer of the same analysis. They found only a very slight advantage, on the average, in favor of the home- mixed fertilizers. In table 21 is shown the relative extent to which home mixing is practiced in the regions surveyed. These figures indicate that an average of about TABLE 21. PER CENT OF GROWERS USING HOME-MIXED FERTILIZER ON POTATOES Long Island Steuben County Monroe County Franklin and Clinton Counties 6 5 10 1 5.5 per cent of the potato growers in New York mix their own fertilizer for potatoes. A comparison of the various ingredients which constitute the home-mixed fertilizers used in these four regions is given in table 22: TABLE 22. PER CENT OF GROWERS USING VARIOUS INGREDIENTS IN HOME-MIXED FERTILIZERS Ingredient Long Island, 1912 Steuben County, . 1912 Monroe County, 1913 Franklin and Clinton Counties, 1913 Nitrate of soda 95 26 3 100 Acid phosphate . . . 90 84 100 75 JMuriate of potash . . 95 62 87 100 Tankage . . .95 0 10 o Blood and tankage 5 0 13 o Dried blood . . 10 0 10 o Bone meal . . 10 5 0 25 Nitrate of soda and dried blood .... 0 60 10 0 Sulfate of potash 5 21 13 o Raw rock phosphate 10 5 0 0 Fish scrap 55 0 0 0 1180 EARLE V. HARDENBURG The percentages shown in table 22 indicate that on Long Island, nitrate of soda and tankage are used almost universally as the sources of nitrogen in home-mixed fertilizers. As is typical of coast regions, much fish scrap also is used for its nitrogen content. Acid phosphate is the principal source of phosphoric acid, altho ten per cent of the growers who mix their own fertilizers on Long Island use bone meal, and an equal number use raw rock phosphate, for the phosphoric acid supply. Of the potash supply 95 per cent comes from muriate of potash, and the remaining 5 per cent comes in the sulfate form. In Steuben County, nitrate of soda and dried blood used together was the main source of nitrogen, no tankage being used by the five per cent of growers who mixed their own fertilizer. Phosphoric acid was obtained by eighty-four per cent of these growers from acid phosphate, and five per cent obtained it from bone meal. More sulfate of potash was used in this region than in any of the other regions surveyed, altho 62 per cent of all the potash was obtained in the muriate form. One-tenth of the growers visited in Monroe County mix their own fertilizer. About an equal number of these growers obtained their nitro- gen supply from dried blood and from tankage. A few used nitrate of soda. All of these growers obtained their phosphoric acid from acid phosphate. Muriate of potash was used by eighty-seven per cent of the growers, v/hile thirteen per cent used the sulfate form for potash. Only one per cent of the growers in Franklin and Clinton Counties practiced home mixing. These men used only nitrate of soda for nitrogen and muriate of potash for potash. Acid phosphate was the principal source of phosphoric acid, altho a little bone meal was used. Summarizing for the four regions, it is seen that nitrate of soda is the commonest source of nitrogen. Generally, however, there is a tendency to mix nitrate of soda with either dried blood or tankage to furnish nitrogen in both a quickly and a slowly available form. Acid phosphate and muriate of potash are the principal sources of phosphoric acid and potash, respectively. Quantity and value of fertilizer The optimum amount of fertilizer for any crop necessarily depends on three principal factors: the available supply of plant food in the soil, the feeding requirements of the crop, and the net return per unit invested in fertilizer. Of these, the first two are usually measured by the yield per acre, regardless of cost, while the last is too often neglected. Macoun (1905) has shown, by a large number of analyses, that a 200-bushel yield of potatoes (exclusive of the tops) removes an average of 40 pounds of nitrogen, 20 pounds of phosphoric acid, and 70 pounds of potash, per acre of soil. This is about the same amount of nitrogen and phosphoric acid, but twice as much potash, as is removed by comparable yields of wheat and corn. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1181 Whitney (1910) summarized the many fertilizer experiments on potatoes in this country as to the influence of increasing the amount of fertilizer. So far as nitrate of soda, acid phosphate, and muriate of potash, used singly, are concerned ^no consistent increase in yield has resulted from increasing the amount up to 500 pounds per acre. Increasing the amount of complete commercial fertilizer up to a ton and more per acre has, on the contrary, consistently increased the yields. The figures shown, however, would indicate that the increase in yield caused by amounts exceeding a ton has not been profitable. The same conclusions can be drawn regarding the use of manure in these experiments up to an amount not exceeding 20 tons per acre. One of the oldest and best series of fertilizer experiments on potatoes, covering many years, was begun on Long Island in 1895, by the New York Agricultural Experiment Station at Geneva (Van Slyke, 1895). Com- parison of the yields from the use of 500, 1000, 1500, and 2000 pounds of fertilizer, up to 1898, showed that it was not profitable to use more than 1000 pounds per acre. The tests in 1898 showed 1500 pounds to be the most profitable amount. The curve of relation between the cost of fer- tilizer and the yield of potatoes, altho irregular, shows a positive correla- tion. Jordan (1900), reporting on a continuation of these Long Island tests in 1900, showed that whereas the highest yields were obtained with 2000 pounds of fertilizer per acre, the greatest net gain from the crop resulted when only 1000 pounds was used. Rane and Hall (1904), at the New Hampshire station, found that 1500 pounds of commercial fertilizer was the most profitable amount to use, whether or not normal applica- tions of manure were used. Greater amounts of fertilizer, either with or without manure, were not profitable. Kohler (1910), in a triplicate series of plots conducted at the Minnesota station in 1910, showed that under Minnesota conditions it would not pay to use more than 800 pounds of fertilizer per acre, and in most of his tests 650 pounds gave the highest gain. The gain in yield from the elements used singly was almost negli- gible, their efficiency showing only when in combination. Kohler recom- mended the use of commercial fertilizer only 'when the supply of stable manure became insufficient in quantity. The experiments of Zavitz (1916) at the Ontario station, covering cooperative and station tests for five and three years, respectively, show a gain in yield, in most cases, resulting from an increase in either the amount or the value of the ferti- lizer used. Manure and fertilizer in combination, and manure alone, gave the greatest yields per acre and formed the cheapest fertilizer in both sets of experiments, not counting the cost of freight and application. So far as profit is concerned, therefore, the results of the Ontario experi- ments must be discounted. General experience has shown $h&t the high cost of handling stable manure for potatoes on a large scale is often prohibitive. 1182 EARLE V. HARDENBURG Manure and fertilizer used in the four regions There is considerable variation in the amount of manure and fertilizer used in the four regions surveyed. On Long Island, where the crop is grown successively on the same land, it is necessary to use large amounts of fertilizer in order to maintain the yields. In table 23 are given data concerning the use of manure and fertilizer in the four areas surveyed. It is obvious from this table that the use of manure on potatoes is closely asso- ciated with, and largely dependent upon, the dairy industry. TABLE 23. SUMMARY OF MANURE AND FERTILIZER USED IN THE FOUR REGIONS SURVEYED Average Per cent value Per cent Average Per cent Average of per acre of amount of amount growers of manure growers of growers of Region using and using fertilizer using manure manure or fertilizer fertilizer per manure per fertilizer on farms on acre on acre or both using potatoes (pounds) potatoes (tons) them Long Island . 100 $32.42 100 1,922 21 5.2 Steuben County .... 95 11.00 39 270 93 12.2 Monroe County. . . . 100 14.84 65 406 98 12.0 Franklin and Clinton Counties 99 13.14 76 516 79 11.0 The first column of percentages .in table 23 includes not only the growers who applied manure or fertilizer directly to potatoes, but also those who applied manure or fertilizer to the crop preceding potatoes, the potatoes receiving a certain percentage of value from the residue. On this basis, the average value of manure and fertilizer used on Long Island was more than twice that for the Monroe and the Franklin and Clinton County areas, and nearly three times that for the Steuben County region. The second and third columns of percentages in the table represent the growers who applied fertilizer and those who applied manure, respectively, directly to the potato crop. (The reader is referred to page 1178. for the method used in, evaluating manure.) It may be noted that manure is used directly for the potato crop by almost every grower in Steuben and Monroe Counties, while on Long Island only one grower in five uses it in this way. Growers on Long Island do not find it so practicable because of the expense of handling it for large acreages, the danger of scab infection, and the insufficiency of the supply for their fertilizer needs. In Steuben and Monroe Counties, manure is almost invariably applied to the sod land previous to plowing for potatoes. It is so applied also, but to a lesser A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1183 extent, in Franklin and Clinton Counties, altho here much manure is applied to new seeding and some is applied to oats. It is evident from table 23 that the region in which .only a small propor- tion of the growers use fertilizer on potatoes is also the region in which the least fertilizer is applied to an acre. The average application of manure per acre in each region shown in the table, is reckoned not on the basis of those acres alone which received manure, but on the basis of the total potato acreage of those farms where manure was applied. Thus, on Long Island, manure is applied to only a small proportion of the total potato acreage per farm, while in the other regions most of the acreage is covered. The amount per acre averages nearly 10 tons, tho the rate varies from 10 to 20 tons. The extent to which fertilizer and manure are used in Monroe County and in Franklin and Clinton Counties is fairly similar. The least fertilizer is used in Steuben County. Whether more could be used profitably in any of these regions is discussed in the subsequent studies. Value of manure and fertilizer on Long Island Inasmuch as the amount of seed used has been found to be a very influential factor in determining yield, this factor is eliminated as far as possible in the studies of other factors. Therefore, in studying the influence of the value of manure and fertilizer on yield, the records were first sorted into groups, according to the value of manure and fertilizer, and were then resorted according to the amount of seed used, as shown in table 24: TABLE 24. RELATION OF VALUE OF MANURE AND FERTILIZER TO YIELD ON 330 LONG ISLAND FARMS IN 1912 Less than 12 From 12 to 14 14 bushels and bushels of seed bushels of seed more of seed Average of totals per acre per acre per acre Value of manure and fertilizer per acre Num- Average yield Num- Average yield Num- Average yield Num- Average yield ber of per ber of per ber of per ber of per farms acre farms acre farms acre farms acre (bushels) (bushels) (bushels) (bushels) Less than $30 . . 67 155.5 43 157.1 13 176.4 123 158.9 $30-$40 58 172.1 60 174.0 38 196.2 156 180.1 $40 and over 14 182.7 23 187.5 14 222.1 51 198.0 Total 139 126 65 330 Average . 166.0 170.2 197.8 175.5 1184 EARLE V. HAEDENBURG With the amount of seed used remaining constant, the yield was increased in every instance by an increase in the value of manure and fertilizer used. Furthermore, the yields were apparently sufficiently increased by the use of fertilizer to the value of at least $40 an acre, to make such applications profitable. It is evident that the maximum limit of fertilization in 1912 did not exceed the point of optimum profit. The correlation between value of manure and fertilizer, and yield per acre, for this region is significantly expressed by the positive coefficient 0.244 d= 0.035 shown in figure 136. S £ Yield per acre, in bushels O & lo ^ i § £ -OQ*OO»OQ M i— i § £ i 1 ^ i £ ll $ 1-$ 5 $ 6-S10 2 2 7 9 3 13 4 21 2 23 1 13 6 1 3 H * $1 1-115 5 3 21 15 38 14 8 5 2 ll $16-$20 1 6 8 10 6 7 9 3 2 05 " O FH $21-$25 1 1 2 1 3 2 1 1 1 ^| $26-$30 1 1 1 2 1 1 1 "§ $31-$35 1 1, $36-$40 2 1 A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1189 fertilizers; (3) the use of organic fertilizers in combination with minerals resulted in higher average yields than resulted from mineral fertilizers alone; and (4) manure and commercial fertilizers produced higher average yields than any other type of fertilizer. Ballou (1910) and Gourley (1910) substantiate these general conclusions in reporting the results of a fifteen- years comparison of nitrate of soda, acid phosphate, and muriate of potash, used both alone and in combination on potatoes. These tests have shown that, altho the cheapest cost of increase per bushel was obtained from the use of acid phosphate alone, the greatest profit per acre resulted from the use of the complete fertilizer. The Rothamsted station, in England (Hall, 1905), experimenting for twenty-six years and using five varieties of potatoes on a series of ten plots, compared the yields from plots receiv- ing ammonium salts alone, nitrate of soda alone, superphosphate alone, and mixed mineral fertilizer. The average yields resulting from these treatments varied in the order listed, ranging from the lowest yields with ammonium salts to the highest yields with mixed mineral fertilizer. Since these plots grew potatoes successively for twenty-six years, and since potatoes yielded less with nitrogenous fertilizers than with mineral salts alone, it was concluded that " the potato finds a difficulty in obtaining ash constituents rather than nitrogen from an impoverished soil." Balen- tine (1894) conducted greenhouse experiments to compare the foraging power of the potato plant for phosphoric acid, with that of other crop plants. He used identical amounts and forms of nitrate and potash fertilizer, but varied the phosphatic form. He compared the results from equivalent amounts of phosphoric acid in mostly insoluble forms with those from the soluble form and with the check in each crop series. The results showed that the potato plant is not able to make use of this element in the insoluble form nearly so well as do wheat, corn, peas, and turnips. Because of its importance in stimulating the vigor and yield in the crop and in satisfying the feeding requirements of the plant, potash had occupied the most important place in potato fertilizers up to the time when this survey was made. Rane and Hall (1904) compared the yields from plots containing 5, 10, and 15 per cent of potash, respectively, and those from plots containing no potash at all. They found that at the New Hampshire station, altho the yield was increased up to 15 per cent of potash, the most profitable results came from the 10 per cent of -potash in a complete fertilizer. Several years later T. C. Johnson (1916) com- pared complete fertilizers differing only in that they contained 3, 5, and 7 per cent, respectively, of potash. He obtained the best results from the 5-per cent fertilizer, since that containing 7 per cent of potash seemed to retard maturity and decrease the yield. Conner (1906), at the Florida station, compared complete fertilizers containing 7, 8, 9, and 10 per cent of potash, respectively. Tho this was but a one-year test, the check plots averaged nearly as high yields as did the plots receiving potash, 1190 EARLE V. HARDENBURG which indicates 'that probably potash ' is not a limiting factor in the soil at the Florida station. Under the recent war conditions, growers in the Eastern States were forced to do without potash or to use less. Woods (1918), at the Maine station, has attempted to determine the possibility of obtaining satis- factory yields without this element. His average results for the three years 1915, 1916, and 1917 show that, whereas fairly good yields have been obtained with no potash, the yield has been increased 26 bushels an acre by the addition of 3 per cent of potash. The additions of 5 and 8 per cent of potash have given practically no increase above that from 3 per cent. The high percentage of potash previously used in Maine was evidently not needed, or else the soil had become temporarily stocked with a surplus. Chemical analyses have shown that the sandier soils of the coastal plain are more deficient in potash than the heavier soils farther inland. This fact and the more intensive cultivation of potatoes probably account for the larger amounts of potash previously used in these regions. It might be supposed, therefore, that yields of potatoes cannot long be maintained without this constituent. The recent studies of Dr. Oswald Schreiner, of the United States Department of Agriculture, on potash hunger in the Aroostook potato region in Maine and in the Norfolk potato truck areas, bear out this conclusion. The writer has recently observed marked examples of potash hunger in the potato fields of Long Island. Evidently the shortage of potash is beginning to be felt. Of the two principal forms of potash — muriate, or the chloride form, and sulfate — the muriate has always been the more commonly used on this crop. The reasons for this are the greater cost and the lesser supply of the sulfate form. It is occasionally stated, tho the point does not seem to be borne out by much experimental evidence, that the chlorine in muriate of potash is detrimental to quality in the potato. There has generally been little difference between the two forms as to the yield resulting. Rane and Hunt (1897), in a one-year test, used 87 varieties of potatoes and obtained a very slight advantage in favor of muriate of potash. Many years later, Brooks (1914) obtained an increase in yield of 11 bushels per acre in favor of the sulfate form, with an additional improve- ment in quality. The experiments of Conner (1906) show results from the use of various amounts of ammonia and of phosphoric acid in complete fertilizer, which not only are inconsistent but also show average yields not essentially different from those from the check plots. T. C. Johnson (1916), com- paring the results from 2, 4, and 8 per cent, respectively, of phosphoric acid in complete fertilizer, found the best results from the 8-per-cent analysis. His results with varying amounts of nitrogen were incon- clusive. Woods (1918), in a test to compare nitrogenous fertilizers in combinations of nitrate of soda, ammonium sulfate, and organic forms, A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1191 thru the years 1914 to 1917, inclusive, found practically no advantage for any one form over another. Altho many different analyses of fertilizers were used in 1912 and. 1913 in the surveyed regions, the majority of growers in each region were buying fertilizers of the same analysis. In table 28 are shown the analyses of the fertilizers in commonest use at that time, in the four regions: TABLE 28. COMMONEST ANALYSES OF FERTILIZERS USED IN THE FOUR REGIONS SURVEYED Rank Franklin Long Steuben Monroe and Island, County, County, Clinton 1912 1912 1913 Counties, • 1913 Used by majority 3-8-7 2-8-10 2-8-10 2-8-10 Second in amount used 4-8-7 0-10-8 0-10-8 0-10-8 Third in amount used 5-8-5 0-8-10 0-8-10 0-8-10 Because of the higher percentage of nitrogen in the fertilizer used on Long Island, this region has used a higher grade of fertilizer than any of the other regions. The extra nitrogen has been used to maintain this element in view of the heavy draft on it caused by the continuous potato culture. The greater use of manure and sod residue in the other regions has furnished the nitrogen lacking in the additional fertilizer used. Aside from the nitrogen content, the fertilizers used in the four regions have been similar in analysis. Method of applying fertilizer The method of applying fertilizer to potatoes is a question not well worked out because it is complicated by, and dependent on, such factors as soil type, amount of fertilizer used, and method of planting the crop. The question as treated here relates to (1) the depth of applying the fertilizer relative to the seed pieces, (2) broadcasting as compared with drilling, and (3) time of application relative to time of planting the crop. Obviously, fertilizer should be applied at that depth which will afford it a constant supply of moisture to make it available to the root system of the potato plant. As this ddpth is less in heavy soils than in light soils, both fertilizer and seed are generally applied less deeply in heavy than in light soils. Since the root system of the potato plant normally develops laterally to a radius of from 18 to 24 inches, the question of depth of appli- cation would seem more important than that of whether the application should be by drill or broadcast. The question of time of application must depend on the availability of the fertilizer used and the seasonal distri- bution of rainfall. 1192 EARLE V. HARDENS URG Taft (1892) reported the comparative yields obtained by placing ferti- lizer above and below the seed piece in a year of light rainfall. All plots gave a gain of from 14 to 34 bushels per acre in favor of the applications below the seed piece. Taft and Coryell (1894) reported on the same test covering three years and using complete fertilizer on three varieties. The results were all in favor of the application of fertilizer below the seed piece, the gain being from 25 to 40 bushels per acre. Munson (1894) compared the relative efficiency of applying fertilizer by the Rural-New- Yorker trench system, in which the fertilizer is placed in mellow earth two inches above the seed, with the method of applying it to the surface and harrowing in. Not enough difference in yield was obtained to pay for the extra labor of making the trench required in the former method. Rane and Hall (1904), replicating plots three times, compared yields from fertilizer applied above and below the seed, both with and without the use of stable manure. They obtained in all cases an average difference of 18 bushels per acre in favor of the shallow applications. No mention was made of the rainfall available that year. Van Slyke (1895), at the New York station, compared potato yields grown under 1000, 1500, and 2000 pounds of fertilizer per acre, respec- tively, applied both broadcast and 'in the drill row. In all cases in which 1500 pounds or less was used, the drill-row applications gave the better yield by about 10 bushels per acre. When 2000 pounds was applied, there was a difference of 17.5 bushels per acre in favor of broadcasting. This difference was probably due to a slight injury to the seed pieces caused by contact with the fertilizer applied in the drill row, for the stand in the latter case was rather uneven. Rane and Hall (1904), using 1500 pounds of fertilizer per acre, compared the yields obtained by applying all in the hill with those from applying half in the hill and half broadcast. No check plots, and only three plots altogether, were used, but a small yield favoring the half-and-half method was obtained. The gain, however, was due to a larger yield of culls under this method. Hall (1905), in reporting the experience of the Rothamsted station, stated that phosphoric acid and potash should generally be applied in drills, but that kainit should be applied broadcast. Jordan and Sirrine (1910) compared these two methods of application at three points on Long Island during the years 1905 to 1908, inclusive. Altho the differences were small in all cases, there was an average gain of 3 bushels per acre in favor of the drill method. Applications of 500, 1000, and 1500 pounds, respectively, were compared under each method. Woods (1917), using 1000 pounds and 1500 pounds of fertilizer, respectively, all broadcast, all drilled, and a part used either way with the remainder applied after the crop was up, found differences favoring the drill method but no greater than might be expected as within experimental error. The greatest average difference between the two methods, thru the years 1914 to 1916, inclusive, was 6 barrels in favor A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1193 of drilling. Woods concluded that nothing is to be gained by the practice commonly followed in Maine, of applying some fertilizer along the row at the first cultivation, for this method is less convenient and apparently no more efficient than applying all the fertilizer in the row at planting time. Much variation in the method of applying fertilizer has been observed in the surveyed areas. Wherever machine planters were used, the ferti- lizer was generally applied with these. On Long Island, however, where 98 per cent of. the crop was planted by machine, nearly forty per cent of the growers applied the fertilizer broadcast, using a grain drill or a lime sower before planting. Tables included in the study of this factor show that most of the fertilizer not applied thru the planter was applied broadcast before planting. Generally this means an application only a few days prior to planting the crop. Without exception, on Long Island a higher average yield resulted from the broadcasting of fertilizer before planting (table 29). This difference in favor of broadcasting is in part due to the greater quantity of seed usually planted when this method is used. But the difference is sufficient to be significant. Furthermore, the growers in the first fertilizer group who broadcast the fertilizer, used less seed than did those who applied the fertilizer in the drill row. Apparently the average of nearly a ton of fertilizer per acre used in this region in 1912, was more efficient when broad- cast than when drilled. There are rather too few farms included in each of the groups for Steuben County (table 30) to aUow any definite conclusions to be drawn. The differences in yield apparently due to difference in the method of fertilizer application, are all probably due mainly to the indicated difference in the amount of seed planted. This leads to the conclusion that. for Steuben County, when as little as the average of 270 pounds of fertilizer per acre is used, it is fully as efficient to apply it all in the drill row as to broadcast it either just before or just after planting. Altho the differences in yield between the two methods employed in Monroe County were small, they are consistent thruout (table 31). Usually, less seed was planted per acre by .those broadcasting fertilizer before planting than was used by those in the other group, yet the yields favor the method of broadcasting. The writer is unable to reconcile this conclusion with that for Steuben County. Whether the exception here favoring broadcasting when an average of only 406 pounds of fertilizer per acre was used, is due to seasonal rainfall conditions, is purely conjecture which can be settled only by controlled experiments over a protracted period. A study of table 32, weighing the influence of the amount of seed and the value of the fertilizer used in Franklin and Clinton Counties, indicates that here, as in Monroe County, the broadcasting of the fertilizer proved more efficient in 1913 than the application of it in the drill row. It is of 1194 EARLE V. HARDENBURG I ill ill I O CO COIN §•?•«• Ill' l°*: *££ O5O iH M» ' •2fc ; is ; A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1195 1 l» •oj 3 e III < ° : III oco (NfO OC5 CO(N m ^ I III a"§§ <1 ° e Ill CO>OCO -3 i *.««. Is I o^So OCOlO (N CO i-( or^co CO CO O co com CO CO OS I Hi QJ '. G ^ fc --J3 C3 1196 EARLE V. HARDENBURG interest to note that more than half of the 211 growers listed here applied their fertilizer by the broadcasting method. The study of the methods of applying fertilizer, as treated in this paper, does not allow any definite conclusions to be drawn. The question is apparently one of local application, probably depending on such factors as seasonal rainfall, amount of fertilizer used, and soil type, as previously suggested. Use and influence of lime on potato land Lime has been given little prominence in use either as a plant food or as a soil amendment for potatoes. This is due to the fact that, being an alkaline agent, its use is conducive to the development of common scab (Actinomyces chromogenus) on potato tubers whenever the causal organism is present in the soil or introduced on the seed tubers. Therefore lime is usually applied in the rotation as far removed from the potato crop as is possible, while its major benefits to this crop, in the improvement of soil texture and the growth of legumes, are as well accomplished. Wheeler and Adams (1909) reported an increase in the proportion of tubers of mer- chantable size from the use of lime. There seems to be little or no con- clusive evidence available that lime has increased the yield of potatoes except indirectly thru the benefits just mentioned. In spite of the fact that lime is advised for most potato farms outside of the surveyed areas of Long Island and Monroe County, very few growers reported its use in the rotation which they were using at the time when the survey was made. A summary of the use of lime and its place in the rotation followed in the surveyed regions is given in table 33: TABLE 33. USE OF LIME, AND ITS PLACE IN THE FOUR AREAS SURVEYED Region Per cent of growers using lime in rotation Average number of years removed from potatoes Long Island, 1912 6 3.4 Steuben County, 1912 16 3.1 Monroe County, 1913 16 3.0 Franklin and Clinton Counties, 1913 7 3.7 Very few growers apply lime regularly in their rotation. The commoner practice is to use it only when necessity demands it as a means of producing legumes. On Long Island, where potatoes are grown for so many suc- cessive seasons on the same land, it is unsafe to use lime at all. The number of years from the potato crop that lime is applied, indicated in table 33, shows the consistent effort of the growers to apply it as far from potatoes as is possible, in order to avoid scab. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1197 ADAPTATION AND YIELD OF VARIETIES The crop survey offers one of the best means of determining the relative merits of potato varieties for a given locality. Too often the experiment stations have made generalized recommendations solely on the basis of the performance of a few strains tested for only a few years at the station grounds. A correct knowledge of the adaptability of various types and varieties to given soil and climatic conditions can be obtained only by cooperative controlled tests under varying conditions, or by a crop survey of the performance of the varieties growing over a wide area with diverse condi- tions. Nearly every state experiment station in the United States has at some time conducted a yield test of potato varieties, the results of which are to be found in the published literature. These results are in most cases of very local significance and pertain only to the strains of seed that were available for the test. Because of the wide variation in yield of the dif- ferent strains of a given variety, no absolute recommendations for any variety should be made on the basis of such tests. A comparison by survey methods of the average yields of strains of the varieties within a region, furnishes the best criterion of the merits of such varieties for that region. Stuart (1915) has classified the standard American varieties into groups containing varieties similar in tuber and foliage characters. It is now well known that the varieties within each group conform fairly closely to one another in. their adaptation to specific soil and climatic conditions. This has made it possible to determine the type or group of varieties best adapted to certain regions. It remains, then, only to choose high-yielding strains of standard varieties within this group. The status of varieties within each of the surveyed areas has been studied on this basis. Varieties and types have been tabulated in the order of their extent of production in each region. Potato varieties on Long Island Of the four regions surveyed, Long Island presented the greatest varietal standardization by growing the fewest varieties and the fewest types. Growers in this region are convinced that varieties of the Rural group yield less, are poorer in quality, and are less popular in the New York market, than varieties of the Green Mountain, or white-sprout, type. Generally speaking, for the medium late crop, only varieties of the Green Mountain group are raised on Long Island, and the early varieties for this region are chosen from the Cobbler, the Early Ohio, the Rose, and the Triumph groups (table 34). Altho Green Mountain was only one of several varieties of this group grown in the three years from 1911 to 1913 inclusive, its popularity is shown by the fact that two-thirds of the average total acreage during this period was given to this variety. Irish Cobbler was the leading early variety produced, and most of the acreage of this variety was grown in Nassau 1198 EARLE V. HARDENBURG TABLE 34. VARIETIES GROWN ON 330 LONG ISLAND FARMS IN 1911, 1912, AND 1913 Variety j Color of sprout Average number of farms raising variety in 1911, 1912, and 1913 Average yield per acre in 1911 and 1912 (bushels) Per cent of total acreage grown in 1911, 1912, and 1913 Green Mountain White.. 200.7 180.4 66.0 Irish Cobbler Pink 119.7 169.7 13.0 Carman No 1 White . 34 0 189 3 10 0 Delaware White . . . 19 3 210 5 3 0 Mills Pride . . . White 16 3 164 4 2 0 Uncle Sam ' Blue 9 7 220 4 1 6 Early Ohio . . Pink 9 7 216 9 1.6 Norcross .... White 8.7 212.1 1.0 World's Fair Genesee Seedling Rose * Bliss Triumph Early Rose White White Pink Pink Pink 3.7 5.3 5.7 13.0 7.0 185.8 248.1 172.8 126.9 104.5 0.5 0.3 0.3 0.3 0.2 Bagley White 4.3 126.2 0.2 County. A summary of the varieties belonging to each group, as classi- fied by Stuart (1915), is given in table 35: TABLE 35. SUMMARY or TYPES ON 330 LONG ISLAND FARMS IN 1911 AND 1912 Type Per cent of total acreage in 1911 Per cent of total acreage in 1912 Average yield per acre in 1911 (bushels) Average yield per acre in 1912 (bushels) Average amount of seed used in 1912 (bushels) Average value of manure and fertilizer in 1912 Green Mountain . 86 84 186 6 179 7 12 9 $32.74 Cobbler 10 12 189 0 157 2 12 2 32.00 Triumph 0 1 128 6 11 4 28.71 Early Ohio . 2 1 199 6 231 5 12 3 44.99 Rose 1 1 166 8 123 6 13 4 38.72 Rural ... .... 1 1 227 7 216 2 10.1 27.66 Since the Green Mountain group is the only one of importance in this region, no comparison of relative merits is made between the types. Because most of the Early Ohio acreage was produced near or at Orient Point, under high fertilization and with ideal moisture conditions, its high average yield must be discounted when compared to that of the Cobbler group. Owing to its high average yield and its white skin, Cobbler has proved to be the best early variety for this section. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1199 Potato varieties in Steuben County In contrast to Long Island, Steuben County showed the least standard- ization of varieties of the four regions. The thirty-five varieties listed in table 36 do not represent all that were found in the region, but only those comprising an average of at least 0. 1 per cent of the total acreage for three years. TABLE 36. VARIETIES ON 360 STEUBEN COUNTY FARMS IN 1911, 1912, and 1913 Variety f Color of sprout Average number of farms raising variety in 1911, 1912, and 1913 Average yield per acre in 1911 and 1912 (bushels) Per cent of total acreage grown in 1911, 1912, and 1913 Number 9 Blue 76.7 143.5 22.6 Ruloff White . ... 41 3 126 5 12 3 Spaldings Rose 4 Pink 66 6 139 0 11 2 White Pearl Blue 20 7 148 5 6.6 Sir Walter Raleigh Blue 16.0 156.8 4.7 Gold Coin White . ... 21 0 135 0 4 7 Dooley Blue 15.3 148.8 4.5 Carman No 3 Blue 15 3 149 5 4 2 White Giant Blue 13.0 135.3 4.2 State of Wisconsin White 12 0 134 4 3 1 Rural New Yorker No 2 Blue 7 7 134 4 2 6 Green Mountain White 7 3 115 8 1 8 McKinley Blue 8 0 142 9 1 7 Pan American Blue 4 0 121 3 1 4 Wliite Granger Blue 6 3 147 2 1.4 Pearline Blue 4 0 114 9 1 3 German Queen Pink 4 7 154 0 1 1 Planet Blue 3 7 126 8 10 Million Dollar Blue 3 0 118 7 1 0 Early M^anistee Pink 5 7 163 7 0 9 Uncle Sam Blue 3 0 134 0 0 9 Charles Downing White 3 0 118 2 0 8 American Banner Blue 2 7 127 6 0 8 Goldstein Blue 1 7 118 0 0 7 Admiral Dewey Blue 2 ?'• 167 2 0 7 California White 1 7 135 6 0 7 White Mammoth White 3 0 133 8 0 6 Knoxall Blue 3 7 159 6 0 5 Carlisle Blue 2 0 156 0 0 5 Norcross White 3 3 143 0 0 4 Ward's Seedling Blue 2 7 136 4 0 3 Scotch Mane Pink 2 3 127 0 0 3 Mix Best Blue 2 0 168 9 0 2 Early Burpee Pink 2 0 108 3 0 1 Clustic Beauty White . 0 7 106 2 0.1 1200 EARLE V. HARDENBURG It is significant that the three most popular varieties of this region, repre- senting nearly half of the average total acreage during 1911 to 1913 inclu- sive, are each of a distinct type. Yet each may have its proper place in Steuben County farming. Number 9, representing a high-yielding strain selected from Rural New Yorker No. 2, heads the list in table 36 and is an ideal blue-sprout variety, adapted to the heavy soils and narrow-valley farms of this region. Spaldings Rose 4, a medium early variety of the pink- sprout type, is profitably raised for a special seed trade with the Hastings potato section of Florida . Ruloff is a variety of the Green Mountain type which is well adapted to the lighter soils of the northern part of the county. Altho there may be this apparent justification for diversity of type, there is surely no justification for so many varieties. Among the more popular varieties listed in the table, such standard varieties as Sir Walter Raleigh and Carman No. 3 would, on the basis of yield, appear to justify their more exclusive use in this region. The average yield of the more popular standard white-sprout varieties in this list is considerably inferior to that of the Rural varieties, which apparently indicates that, in general, this type is not so well adapted here as is that represented by Number 9. A comparison of the types produced in this region is shown in table 37: TABLE 37. SUMMARY OF TYPES ON 360 STEUBEN COUNTY FARMS IN 1911 AND 1912 Type Per cent of total acreage in 1911 Per cent of total acreage in 1912 Average yield per acre in 1911 (bushels) Average yield per acre in 1912 (bushels) Average amount of seed used in 1912 (bushels) Average value of manure and fertilizer in 1912 Rural 76 75 138.0 142.4 8.8 $ 8.90 Green Mountain Rose . 14 9 15 9 126.3 113 8 123.1 142 2 15.5 9 7 13.57 8.89 Hebron 1 1 116 7 194 4 9.8 12.82 In both 1911 and 1912, the Rural group of varieties outyielded those of the Green Mountain type by an average difference of 12 and 19 bushels per acre, respectively, in spite of the fact that in both years they were grown with considerably less seed and fertilizer. As is shown later, in table 42, a part of this difference was due to a greater average loss per acre in the unharvested yield, due to late blight rot, in the Green Mountain varieties. It therefore appears certain that the Rural type of potato is better adapted to the prevailing conditions of this region. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1201 Potato varieties in Monroe County Altho a very large number of varieties were grown in Monroe County during the years 1912 to 1914, inclusive, it is seen in table 38 that the TABLE 38. VARIETIES ON 300 MONROE COUNTY FARMS IN 1912, 1913, AND 1914 Variety / Color of sprout Average number of farms raising variety in 1912, 1913, and 1914 Average yield per acre in 1912 and 1913 (bushels) Per cent of total acreage grown in 1912, 1913, and 1914 Sir Walter Raleigh Blue. . 59 3 103 9 20.7 Carman No. 3 Blue 37 3 128 2 13.5 Number 9 . Blue 45 3 125 6 12.3 Peerless Jr. Blue 21 0 124 5 8.3 White Giant . Blue. 22 3 129 9 6.8 Million Dollar . Blue . 21 0 147 0 6.1 Perfection . ... Blue 17 0 144 1 5.5 World's Wonder White Flyers Blue ? 22.3 10 0 138.9 119 2 4.5 2.6 White Grant Blue 14 0 130 2 2.4 Rural New Yorker No. 2 Gold Coin Blue.. White... 7.7 3 3 124.1 111 0 2.3 1.6 American Giant White . 6 3 133 4 1.4 Granger Blue 5 3 109 2 1.4 Twentieth Century Blue 6 0 164 9 1.3 Isle of Jersey Blue 3 0 110 8 1.2 Hundred Fold ? 4 7 141 2 1.2 Irish Cobbler . . Pink 11 7 124 6 1.2 Number 6 Blue. . 4 7 146 5 1.2 Pan American Blue 4 7 138 5 1.1 McKinley Blue 4 3 156 5 0.9 Green Mountain White .. 3 7 174 9 0.8 Early Michigan. Pink 2 0 115 0 0.5 Livingston Pink 2 0 90 6 0 5 American Banner Blue 2.0 153.7 0.4 Number 8 Blue > -.'— 2'0 150 8 0.3 prevailing type was that of the Rural group. Nearly half of the average total acreage of this region during the years 1912 to 1914, inclusive, was planted to the three standard blue-sprout varieties, Sir Walter Raleigh, Carman No. 3, and Number 9. Altho the three-years average yield for these varieties was slightly less than the average yield for the region in 1913, it would still seem desirable, for the sake of standardization, to select high- yielding strains from, and to retain, these few varieties to the exclusion of most of the other varieties of the Rural type listed in table 38. It is note- worthy that of the twenty-six varieties listed, only three are of the Green 1202 EARLE V. HARDENBURG Mountain type. The relatively light seasonal rainfall and the heavy soils of the Dunkirk and Ontario series have resulted in the survival of Rural varieties at the expense of other types. Here, as on Long Island, Irish Cobbler was found to be the leading early variety. The status of varietal types in this region is summarized in table 39 : TABLE 39. SUMMARY OF TYPES ON 300 MONROE COUNTY FARMS IN 1912 AND 1913 Type Per cent of total acreage in 1912 Per cent of total acreage in 1913 Average yield per acre in 1912 (bushels) Average yield per acre in 1913 (bushels) Average amount of seed used in 1913 (bushels) Average value of manure and fertilizer in 1913 Rural 94.0 94.0 141 4 125.0 12.8 $11.48 Green Mountain 3.4 3.0 153.3 109.4 14.3 10.97 Cobbler 1 0 1 0 135 9 117 1 11 6 12.75 Hebrcn 1 0 0 5 121 0 138 8 14 5 9.65 Rose 0 3 0 S 128 6 80 0 10 6 4.40 Early Michigan 0 3 0 7 121 4 100 0 11 1 5.56 The year 1913 was one of low seasonal rainfall in Monroe County, and the crop suffered from the drought. As shown in table 39, varieties of the Green Mountain group yielded, in that year, an average of about 15 bushels per acre less than those of the Rural type, in spite of the use of more seed and about the same amount of fertilizer. This is further evi- dence that varieties of the Green Mountain group, which set tubers earlier than do those of the Rural group, suffer the effects of reaching their critical growth period during the drought season of midsummer in Monroe County. Potato varieties in Franklin and Clinton Counties Altho more varieties are grown in Franklin and Clinton Counties than on Long Island, this region is similar to the lLong Island area in that nearly all the varieties were found to be of the Green Mountain type, as is shown in table 40. A notable absence of early varieties was found among those grown from 1912 to 1914, inclusive. This may be explained in general by the fact that the possibilities for profitable yields are much greater in the main crop varieties. The four Green Mountain varieties listed at the head of table 40 — White Beauty, Selfic Beauty, World's Fair, and Immense — comprised more than half of the average total acreage of this region during the three years con- cerned. Type has become well standardized here, as the list shows only four varieties of the Rural, or blue-sprout type, the remainder all being of the Green Mountain, or white-sprout, type. While these four Rural varieties are standard varieties, their average yield per acre and extent of A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1203 TABLE 40. VARIETIES ON 300 FRANKLIN AND CLINTON COUNTY FARMS IN 1912, 1913, AND 1914 Variety Color of sprout Average number of farms raising variety in 1912, 1913, and 1914 Average yield per acre in 1912 and 1913 (bushels) Per cent of total acreage grown in 1912, 1913, and 1914 White Beauty White 47.3 198.1 16.4 Selfic Beauty White 39.3 200.7 15.0 World's Fair White 32.3 179.9 12.3 Immense White 17.7. 154.5 7.1 White Mountain White 20.3 199.1 6.7 Green Mountain White 18.7 179.5 6.6 White 14 0 176.8 4.7 White Lady White 8.7 197.9 4.1 Number Q Blue 11.3 168.9 4.0 Mill's Prize ' White 7.7 189.9 3.0 Carman No. 1 National White White 10.7 6.7 204.5 184.8 3.0 2.4 Rural New Yorker No 2 Blue 6.3 135.6 2.0 Lincoln White 6.3 221.8 1.9 Scott ? 4.7 160.6 1.8 Mountain King White . . . 5.3 199.7 1.7 Eggswaire White . 5.0 165.6 1.7 Carman No. 2 . Blue 5.3 191.9 1.6 Success ? 3.7 169.9 1.2 International White 3.0 197.3 0.9 Gold Coin White 2.0 147.7 0.7 Silver Dollar Carman No. 3 White Blue 4.7 1.7 187.4 184.0 0.7 0.5 production do not seem to justify their competition with the Green Moun- tain type A comparison of these two types for the region in 1912 and 1913 is shown in table 41. The averages for 1912 and 1913 show that with TABLE 41. SUMMARY OF TYPES ON 300 FRANKLIN AND CLINTON COUNTY FARMS IN 1912 AND 1913 Type Per cent of total acreage in 1912 Per cent of total acreage in 1913 Average yield per acre in 1912 (bushels) Average yield per acre in '1913 (bushels) Average amount of seed used in 1913 (bushels) Average value of manure and fertilizer in 1913 Green Mountain 86 92 194.5 183.8 12.0 $13.44 Rural . 14 8 168.9 164.9 13.1 14.42 1204 EARLE V. HARDENBURG ttl II 6° » | -d 3 ll 00 (N 1C b- 00 T}< t^CO •^ (N (N to iffllN (OCO CO N i-H r-l 21 S ?s £ O oa cc I-H >S^sfl -O t^» CO C^l *O t^» CO C^l »o o WHEELER, H. J., AND ADAMS, G. E. Further results in a rotation of potatoes, rye, and clover. Rhode Island Agr. Exp. Sta. Bul. 135: 99-126. 1909. A STUDY OF FACTORS INFLUENCING THE YIELD OF POTATOES 1279 WHEELER, H. J., To WAR, J. D., AND TUCKER, G. M. The effect of liming upon the development of potato tubers. In Fertilizers. Potatoes. Potato scab. Rhode Island Agr. Exp. Sta. Bui. 33: 46-50. 1895. WHIPPLE, O. B. Thinning experiments with potatoes. Montana Agr. Exp. Sta. Bui. 106 : 1-8. 1915. WHITNEY, MILTON. Fertilizers for potato soils. U. S. Soils Bur. Bui. 65:1-19. 1910. WOODS, CHARLES D. Field experiments. Maine Agr. Exp. Sta. Bui. 188:25-32. 1911. 1914. Field experiments. Maine Agr. Exp. Sta. Bui. 224:25-48. Barn and field experiments in 1916. Maine Agr. Exp. Sta. Bui. 260:85-120. 1917. Barn and field experiments in 1917. Maine Agr. Exp. Sta. Bui. 269:1^4. 1918. WOODS, CHARLES D., AND BARTLETT, J. M. Field experiments in 1906-8. Maine Agr. Exp. Sta. Bui. 167:85-104. 1909. ZAVITZ, C. A. Potatoes. Ontario Agr. Dept. Bui. 239:1-88. 1916. Memoir 52, Studies in Pollen, with Special Reference to Longevity, the fifth precedes: number in this series of publications, was mailed on March 9, 1922. NEW YORK STATE COLLEGE OF AGRICULTURE CORNELL UNIVERSITY, ITHACA, N. Y. DEPARTMENT OF FARM PRACTICE AND FARM CROPS Farm No Potato Record for 1912.. Date , 1913 Operator Age P. O County Location Miles to shipping point Soil types Elevation Topography of farm Drainage Exposure 1912 Acres farmed Value of land per acre ....Acres potatoes -1912 Tenure _ Years owner Years renter Rotation 1 .2 3 4 5 : 6 CROP PRODUCTION FOR 1912 Crop Acres Yield per acre Total Corn, grain... Com silage tons •. bu tons Corn, other tons.... tons... Wheat bu ". bu Rye ^ bu : bu Oats bu bu Barley .' bu • bu Buckwheat bu .:. ; bu Hay tons .... tons.... Alfalfa -. tons. ... tons.... Oat Hay .*. tons.... ,„, tons.... Oats and Barley ....'. tons.... tons.... Oats and Peas : '. tons... .tons,... Field Beans '. bu ........bu Cabbage tons.... ...^ • tons.... Cauliflower ; .-...'. bbl .'.....bbl Brussels Sprouts crates.. .-. crates.. Apples Bearing bu bu Apples not Bearing J POTATO PRODUCTION FOR 1911, 1912 AND 1913. Yew Varieties Acres Yield per acre Total yield Early 1911 Late Total Early :..... 1912 Late .....* : ToUl Early .... . 1913 '. - - Late - Total ; • -v^-^--^ DISPOSAL OF 1912 CROP Date Bushels Price per bu. Total Sold $ $ Seed Feed Home use Total (Ave.) SPECIAL EQUIPMENT Cost Value 1912 Life Depredation Cost of repairs Planter ...$ ...$ ...$ ...$ Sprayer Digger..... Cutter Other equip. _ Total EXPENSES 1912 Price per unit Total Rental value of land Fertilizer Manure from preceding crop.. Manure, used by 1912 crop .. Seed, farm and bought Dust for cut seed Copper sulphate Lime (form for spray) Insecticide (kind) Carriers not returned Equipment Rented Rental value of storage Repairs on machinery Depreciation on machinery Man labor Horse labor Equipment labor .tons... .bu. ... hrs. hre.. Total SUMMARY Total receipts S . . Total expenses .S Profit 1912 crop $ Profit per acre $ LABOR ITEMS Date Acres Per HI. acre To h. j m. M h. Manuring..... Plowing, Spring Plowing, Fall ! Dragging, times „ Discing, times Rolling, times. Cutting seed Treating seed Removing sprouts, times Starting sprouts x; Marking Planting, machine.. Planting, hand Fertilizer..™ Recovering, times Weeding, times Cultivating, times '. Hilling, times - Spraying, times Digging and picking up, (hand) Digging and picking up, (machine) Harrowing after digging Sorting and bagging in cellar Hauling to storage Hauling to market Hauling from storage to market Work on equipment Work on storage .\ Total "MISCELLANEOUS FACTORS Manure. Kind used Tons per acre Source, if bought Where applied Name of spreader, if used often in rotation Piled or spread Plowed in or harrowed Value of residual manure on potatoes, 1912 per cent Plowing. Depth of plowing .^ Seed. Source Amt. used Seed Treatment. Corrosive sublimate. Formalin. Formaldehyde gas. Flowers of sulphur. Formula How treated •• Satisfactory ? Consequent injury to vitality ? Starting Sprouts. Increase in yield noted Increased earliness noted Cutting Seed. Amt. cut Name of the cutter, ;f any Satisfactory ? Type of seed planted : 1. Small whole 2. Medium whole 3. Large cut, 3 or 4 pieces 4. Medium cut, 2 or 3 pieces 5. No. eyes to piece Formula Source of P Ingredients used How long cut before planted Fertilizer. Amt. per acre Source of N Home mixed Cut seed dusted Brand Source of K Amt .Price : DISPOSAL OF 1912 CROP Date Bushels Price per hu. Total Sold $ ...$ Seed Feed Home use Total _ (Ave.) :. SPECIAL EQUIPMENT Cost Value 1912 Life Depreciation Cost of repairs Planter ...$ ....$ ....$ ...$ Sprayer , Digger..... Cutter Other equip _ Total EXPENSES 1912 Amount Price per unit Total Rental value of land acres.. ..$ Fertilizer tons.... Manure from preceding crop tons.... f Manure, used by 1912 crop tons.... Seed, farm and bought ... .bu Dust for cut seed Ibs Copper sulphate Ibs Lime (form for spray) Ibs. Insecticide (kind) Ibs. Carriers not returned Equipment Rented .'. Rental value of storage Repairs on machinery Depreciation on machinery Man labor hrs Horse labor ..- hrs Equipment labor hrs ; Total , SUMMARY Total receipts $ Total expenses $.. Profit 1912 crop $ Profit per acre $ LABOR ITEMS Date Acres Per in. acre To h. | m. tal Manuring..... Plowing. Spring Plowing, Fall Dragging, times „ Discing, times Rolling, times. „ .v. Cutting seed .... Treating seed Removing sprouts, times Starting sprouts Marking „ Planting, machine... Planting, hand Fertilizer...... Recovering, times Weeding, times ". ~i'.i Cultivating, times '. Hilling, times i. Spraying, times Digging and picking up, (hand) Digging and picking up, (machine) Harrowing after digging ;... Sorting and bagging in cellar Hauling to storage Hauling to market Hauling from storage to market Work on equipment Work on storage Total "MISCELLANEOUS FACTORS Manure. Kind used Tons per acre Source, if bought Where applied Name of spreader, if used often in rotation Piled or spread Plowed in or harrowed Value of residual manure on potatoes, 1912 per cent : Plowing. Depth of plowing Seed. Source Amt. used Seed Treatment. Corrosive sublimate. Formalin. Formaldehyde gas. Flowers of sulphur. Formula. How treated .• Satisfactory ? Consequent injury to vitality ? , Starting Sprouts. Increase in yield noted Increased earliness noted Cutting Seed. Amt. cut Name of the cutter, 'f any Satisfactory ? Type of seed planted : 1. Small whole 2. Medium whole 3. Large cut, 3 or 4 pieces 4. Medium cut, 2 or 3 pieces 5. No. eyes to piece : Cut seed dusted Formula Brand Co. Source of P '... Source of K How How long cut before planted Fertilizer. Amt. per acre Source of N Home mixed ... Ingredients used Amt. Price. UNIVERSITY OF CALIFORNIA LIBRARY