COOPERATIVE IMPROVEMENT OF CITRUS VARIETIES By A. D. Shamel Separate No. 813, from Yearbook of the U.S.D.A. 1919. THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA LOS ANGELES ;ORTI< COLLEGE OF AGRICULTURE BERKELEY, CAI IFORNIA IMPROVEMENT CITRUS By A. D. SHAMEL, Physiologist in Charge of Fruit-Im- provement Investigations, Office of Jlorticnltiiral ami PomolOffiCttl Inves- tigations Bureau of Plant Industry. IMPORTANCE OF THE CALIFORNIA CITRUS VARIETIES. DURING the crop year 1918-19 about 39,100 carloads of oranges and grapefruit and 9,963 carloads of lemons, or 49,063 carloads of citrus fruit, were shipped from Cali- fornia. In the wholesale markets this crop brought more than $100,000,000. After the expenses incurred in packing, transporting, and marketing were deducted from this amount, approximately $75,000,000 was returned to the cit- rus growers in California. 1 Only a few varieties of cftrus fruits were grown to pro- duce this result. The Washington Navel orange crop, which ripens during the winter months, amounted to ap- proximately 17,000 carloads; the Valencia orange crop, which ripens during the summer months, amounted to about 20,000 carloads. Other orange varieties of minor impor- tance produced about 1,500 carloads. The Eureka and Lis- bon varieties of lemons, differing mainly in the season of production, so that mature fruits are marketed during the entire year, produced 9,963 carloads. The Marsh is the only grapefruit variety grown commercially, and this crop, which 1 Information furnished by the California Fruit Growers' Exchange. Separate No. 813, from Yearbook of the Department of Agriculture, 1919. 249 172146 20 1 Of 250 Yearbook of the Department of Agriculture, 1910. ripens during the summer months, amounted to about 600 carloads of fruit. The high commercial reputation of the California citrus fruits has largely resulted from marketing regular and uni- form supplies of good fruit .produced by a few standard varieties, which are readily identified by the trade and by the consumers. For this reason the importance to the cit- rus industry of conserving and improving these varieties, now that their reputation has become fully established, must be apparent to every thinking person. Most of the development of the citrus industry in Cali- fornia to its present great commercial importance has taken place within the last 25 years. Its rapid growth during this period is one of the marvels of horticulture. The princi- pal markets for the crop are a long distance from where the fruit is grown. The climate, soil, and cultural condi- tions in the citrus districts necessitate constant vigilance and intelligent effort in order to produce successful crops. These and other circumstances have resulted in the develop- ment of many improved cultural and marketing practices, largely by the aid of scientific research, which have proved to be invaluable not only in the profitable growing and marketing of citrus crops in California, but also in the production and marketing of fruit crops in other sections of the United States. OCCURRENCE AND FREQUENCY OF BUD VARIATION. During recent years many California citrus growers have noticed the presence of undesirable trees in their orchards. This condition seemed more apparent in young orchards or those farther removed from the original trees from which the varieties developed than in "the older orchards which were more closely related to the original parent trees. . Many of these trees apparently produced irregular, light crops of inferior quality. In some cases the commercial and eating quality of the fruit from the offtype trees proved to be so poor that it became necessary to sort them out from the general crop and throw them into the cull bins. This con- dition increased the expense of assorting the crop and also materially reduced the merchantable yield of the orchards. In many instances the inferior and worthless fruits from the undesirable trees could not be easily identified in the Cooperative Improvement of Cities Varieties. 251 packing houses and were inadvertently included with the regular pack; the consumer was disappointed in their eating quality, and a loss of reputation for the crop as a whole inevitably resulted. Acting upon the request of some of the leading citrus growers in southern California, the Bureau of Plant In- dustry in 1909 began an investigation for the purpose (1) of ascertaining the variations which have taken place in the important commercial citrus varieties grown in Cali- fornia through bud variations and to learn the comparative value of the different strains arising from these variations for commercial fruit production; (2) to determine the ex- tent to which undesirable variations have been propagated, as shown by the percentage of such undesirable trees exist- ing in the parent bearing orchards; and (3) through im- proved methods of propagation to reduce the number of undesirable variations which enter into commercial citrus- fruit orchards. These investigations have been carried on by means of records and observations of individual trees. The term "performance record" is used here to mean the record of the number and commercial quality of fruits borne by indi- vidual trees during a period of years. Mostly these studies were made in performance-record plats, consisting of groups of trees gr&wn under comparable conditions, selected for the purpose of determining the behavior of the trees by means of individual-tree records of production, observations, descriptive notes, and photographs. In addition to securing accurate individual-tree yield records, a very careful study of the tree, flower, and fruit characteristics was made. These data showed that striking bud variations were of frequent occurrence in many of the citrus trees. For example, typical Washington Navel orange trees each grown from a single bud and bearing fruits similar to those shown in figure 1, were often found bearing several distinct types of navel oranges, such as those shown in figure 2. In some cases these variations occurred as single fruits possessing characteristics different from those of the Washington Navel orange. In other instances Washington Navel orange trees were discovered having one large Ijmb bearing many fruits which were so different from the other fruits on the tree as to-be classed as belonging to 210105 252 Yearbook of the Department of Agriculture, 1919. a totally different variety. Several individual trees were found on which nine different strains of the navel orange were borne on different limbs, all of them arising as bud sports in these trees. The differences in the characteristics of the fruit variations in some of the trees were found to be almost as important from the commercial standpoint as those which differentiate horticultural varieties. These Best Strain of the Washington Navel Orange. FiG. 1. Typical fruits from a tree of the best strain of the Washington Navel orange variety. variations were not confined to the Washington Xavel orange, but were found almost as frequently in the trees of the other varieties studied. The number of the important fruit variations borne by individual citrus trees differed greatly. A few trees in all of the varieties have been found without any apparent or marked variation in fruits other than the usual modifica- tions of size, shape, texture of rind, color, and quality which are probably due to the influence of season, culture, or other Cooperative Improvement of Citi^us Varieties. 253 environmental conditions. Fortunately, from the stand- point of the conservation and improvement of the varieties, the inherent variations have been found to occur most fre- quently in the trees of the inferior strains of all of the varie- ties. The most productive trees in all cases so far studied and those bearing the most desirable fruits have produced comparatively few of these marked fruit or other bud varia- tions. 1 Dry Strain of the Washington Navel Orange. FIG. 2. Typical fruits from a tree of the dry strain of the Washington Navel orange variety,' showing the undesirable and worthless characteristics of the fruits of this strain. This and other inferior strains originated as bud varia- tions of the best strain, and their accidental propagation has been the source of great loss annually to the growers possessing them. ORIGIN OF STRAINS. The term " strain " is here used to designate a group of individuals of a horticultural variety which differ from all other individuals of the variety in one or more constant and recognizable characteristics capable of perpetuation through vegetative propagation. 1 The detailed results of the investigation of the variations of citrus varie- ties in California have been presented in a series of publications of the United States Department of Agriculture. These reports are in Department Bulletins 623, 624, and 697, to which the reader is referred for further information as to the occurrence and frequency of bud variations. ON OF SUBTROPICAL HORTICULIUKt COLLEGE OF AGRICULTURE BERKELEY. CALIFORNIA 254 Yearbook of the Department of Agriculture, 1919. One of the first and most important individual fruit va- riations observed in Washington Navel orange trees was a large, coarse fruit, which is commonly called an Australian Navel orange. Soon after the discovery of this single fruit variation in the crop of a typical Washington Navel tree, a limb was found in a near-by Washington Navel tree bearing 56 typical Australian fruits. A further study of this Several Strains of Lemons from the Same Tree. FIG. 3. Typical fruits of several strains produced by different branches on a variable Lisbon lemon tree which was grown from a single bud. orchard revealed several trees bearing all, or nearly all, Australian fruits, and having the peculiar upright habit of growth so characteristic of the trees of this strain. An investigation of the single fruit variations found in the trees of the varieties studied revealed their occurrence in other trees as limb sports and in other cases as individual trees. This condition illustrates the probable origin of the Cooperative Improvement of Citnis Varieties. 255 many diverse strains in citrus varieties, due to the accidental propagation of limb sports, and is an important reason for obtaining performance records for use in the selection of bud wood for propagation. So far, 13 strains of the Washington Navel orange, 12 strains of the Valencia orange, 6 strains of the Marsh grapefruit, 8 strains of the Eureka lemon, and 5 strains A Drone Tree. FIG. 4. A typical unproductive or drone tree of the shade-tree strain of the Eureka lemon variety. The trees of this strain show extraordinarily rank vegetative growth and bear light, inferior crops as compared with the trees of the productive strain. of the Lisbon lemon varieties have been found, their char- acterictics described, and the behavior of typical trees de- termined. The origin of all these strains has been traced to bud variations, examples of \vhich are shown in figure 3. Their distribution in established orchards has been largety the result of accidental propagation of the bud variations, due to a lack of knowledge of the importance of the varia- 256 Yearbook of the Department of Agriculture, 1919. tions and their significance in the work of maintaining the citrus varieties. The extent of the occurrence of trees of the diverse strains of the citrus varieties in California has been studied carefully A Productive Lemon Tree. FIG. 5. A typical productive tree of the best strain of the Eureka lemon variety. in many districts by means of orchard surveys. The per- centage of off type trees, that is, trees belonging to strains different from those desired in the orchards and usually inferior to them, has been found to vary from 10 to approxi- mately 90. An average of 25 per cent of the trees in the Cooperative Improvement of Citrus Varieties. 257 orchards studied have been found to belong to strains which differ markedly from the typical or best strain of the variety. The largest percentage of variations from the superior strains has been found in the younger orchards. The trees of one of the strains of the Eureka lemon variety possess unusual vigor of growth and habit- ually bear light crops of infe- rior fruits. A typical tree of this strain is shown in figure 4. On account of their large size as compared with the trees of the productive strain, the density of their foliage, and their poor crops, they have been called shade trees. These trees usu- ally develop a very large number of rank, up- right-g rowing, nonfruiting branches, commonly called suckers. Formerly this sucker growth was generally used by citrus nurserymen for propa- gation. In one of the older Eureka lemon orchards in southern California 10 per cent of the trees were found to be of the Shade-Tree strain. In a younger orchard, the trees of which had been grown from sucker buds secured in the older grove, 25 per cent of the trees were found to be of this strain. In a still younger orchard, where the trees had been grown from sucker buds secured in the second orchard, the percentage of shade trees was found to be 75. This astonish- ing increase in the percentage of shade trees in the younger Orange Variations on the Same Branch. FIG. 6. A branch from a Ruby blood-orange tree bearing a fruit possessing a navel 'and a normal fruit without a navel ; an example of the variation of fruits frequently found in citrus trees. 258 Yearbook of the Department of Agriculture, 1919. orchards is due to the fact that the bud cutters in each in- stance secured a large share of their bud wood from the trees of the Shade-Tree strain, as the sucker growth was most easily secured from them. The shade trees in these three orchards have recently been top-worked, using fruit-bearing bud wood secured from superior trees of the Pro- ductive strain of the Eureka variety, an example of which is shown in figure 5. Without ex- ception the growth and fruits from these selected buds, which were top- work e d on the shade-tree trunks, as shown in figure 7, have proved to be of the de- sired Productive strain, as shown in figure 8. Many other equally striking and im- portant instances of the development and subse- quent elimination of un- desirable strains in Cali- fornia citrus orchards might be described. In every case investigated, the origin of these strains has been traced to bud variations, an example of which is shown in figure 6, which are of frequent occurrence and of very great importance from the viewpoint of the con- servation and improve- ment of the established citrus varieties. A Top- Worked Shade Tree. FIG. 7. A typical Eureka lemon shade tree, such as that shown in figure 4, top- worked with buds secured from a superior performance-record parent tree, such as that shown in figure 5. This photograph was taken three months after top-working. Cooperative Improvement of Citrus Varieties. 259 ISOLATION OF THE STRAINS. Enough evidence has been secured to warrant the asser- tion that all the strains of each of the citrus varieties dis- covered in these investigations can be isolated through bud Good Results from Top- Working. FIG. 8. A top-worked shade tree of the Eureka lemon variety, such as that shown in figure 1, three years after top-working. The barren, rank growth of the original shade tree has been replaced with the productive normal growth of the best strain. Out of 16,000 trees in this orchard 3,200 worthless shade trees have been successfully top-worked. selection. This conclusion is not intended to convey the idea that bud variation within these strains can be entirely eliminated; on the contrary, the investigations have shown that some variation will likely continue as long as the strains are propagated. ftiSiGN OF SUBTROPICAL HORTICULTURE COLLEGE OF AGRICULTURE BERKELEY, CALIFORNIA 260 Yearbook of the Department of Agriculture, 1919. What has been demonstrated is that variation can be con- trolled by bud selection to such an extent that the indi- viduals of the strain can be brought to a condition of prac- tical uniformity as regards crop production and other char- acteristics. At this time there are several thousand acres of bearing citrus orchards in California in which the trees have been propagated from carefully selected buds, secured from superior performance-record trees. In these orchards the progenies of each of the parent trees have been kept sep- arate. Performance-record studies of these progenies and a comparison of their behavior with that of the parent trees conclusively demonstrate that through bud selection each of the important citrus strains has been isolated. The uni- formity of the progenies and the superior and heavy crops of those of desirable strains have proved beyond any doubt that it is practicable commercially to isolate and propagate only the best strains and to eliminate the inferior ones through careful bud selection, l>ased upon individual-tree records and intimate tree knowledge. COMPARATIVE VALUE OF THE STRAINS. As a rule, only one of the many strains in each of the citrus varieties has been found to be worthy of commercial propagation and profitable for cultivation. The value of the product of the trees of the best strains as compared with that of the trees of the inferior strains may be illustrated by the records of production of the trees of the best Washing- ton Navel orange strain and those of the inferior Australian strain in the investigational performance-record plat. In these studies it was found that the trees of the best Washing- ton Navel strain produced an average of 4.73 packed boxes of oranges per tree per year during the period of observa- tion. Under similar conditions the trees of the Australian strain produced 0.76 of a packed box per tree per year. On an acre basis, this yield amounted to 378.6 packed boxes per acre for the best trees, compared with 61 packed boxes per acre for the inferior trees. The actual value of this production was $635.05 per acre annually for the high- producing trees, as compared with $100.0-4 for the low- producing trees of the undesirable strain. Even greater Cooperative Improvement of Citrus Varieties. 261 differences in production and value of the crops from the trees of different strains have been found in other varieties. When it is remembered that on the average 25 per cent of the trees in the orchards studied have been found to be in- ferior strains, the commercial importance of growing only trees of the best strains can be appreciated. OBJECT OF COMMERCIAL TREE-RECORD WORK. The trees of the best strain in e#ch variety have usually been found to be the heaviest producers of fruit. On the other hand, the trees of the inferior strains have usually been found to bear light crops of inferior commercial quality. For this reason individual-tree records of production are of very great value in ^determining the proportion of different strains of trees in citrus orchards. The demonstration of this condition in both experimental and commercial tree- record work has led many of the lead- ing citrus growers to undertake individual-tree record work in their orchards. Such records are now being kept on more than 50,000 acres of citrus orchards in California. The object of the commercial tree records is (1) to locate the drone trees, or those of the inferior strains in the or- chards; (2) to find the superior trees, or those from which bud wood may be secured for propagation; (3) to aid in giving the trees individual care, such as cutting out limb sports or other undesirable growth, treatment for disease, or any tree injury ; and (4) to secure definite evidence as to the effect of cultural treatments and other experimental tests. METHOD OF KEEPING INDIVIDUAL-TREE RECORDS. The method of keeping individual-tree records in citrus orchards now commonly used in California will be briefly described. Various minor modifications of this method have been and are being tried in some orchards, but the prin- ciples underlying this work are fundamentally the same in all cases. It may be found advisable to modify or change the method somewhat, owing to local conditions, but these changes should not be made until experience has shown them to be necessary in order that the records- may be secured in the most natural and logical manner. VISION OF SUBTROPICAL HORTICULTURE * COLLEGE OF AGRICULTURE BERKELEY, CALIFORNIA 262 Yearbook of the Department of Agriculture, 1919. INDIVIDUAL-TREE NUMBERS. Each individual tree in the orchard receives a number. This number consists of three parts, (1) the number of the block or division of the orchard, (2) the number of the row in the block, and (3) the position of the tree in the row. always counting from some fixed point, as, for example, the irrigation head. A tree lo- cated in block 14, row 18, and the twentieth tree in the row, has the number 14-18-20. Where there are several dif- ferent orchards the tree number in the performance- record notes is preceded by the number or name of the orchard or its abbreviation. In the case of bearing trees this number is painted on the tree trunk or on one of the main limbs, arrang- ing the number in a vertical column in the form shown in figure 9. The figures are made with a common letter- ing brush and pure white-lead paint. Very young trees, on which space is not available for painting the number, are designated by attaching a metal or other tag bearing the number. The tree numbers are always placed in the same relative position on all the trees in the orchard, for convenience in finding them. Large, distinct figures are made, so that they are easily legible. The cost of tree numbering has varied somewhat with labor conditions, but at present the numbers are being applied at an average cost of about 2 cents a tree. PICKING. When picking the trees where individual-tree records are secured it is usually necessary to distribute the field boxes to the individual trees instead of in box BOWS, as is ordi- Individual-Tree Numbering. FIG. 9. The arrangement of an in- dividual-tree number on the trunk of a bearing citrus tree in a commer- cial orchard. Cooperative Improvement of Citrus Varieties. 263 narily done. Each picker gathers the fruit from one tree, and usually the same man picks all the trees in each row. All the fruit from each tree is placed in boxes at its base, as shown in figure 10. Care is taken in the beginning to Commercial Performance Record Taking. FIG. 10. Securing performance record of the number of full boxes and the weight of a partly filled box of fruit produced by a Washington Navel orange tree in a commercial orchard. see that none of the fruit from a tree is accidentally carried in the picking sack to a neighboring tree. Pickers quickly realize the importance of keeping the fruit of each tree sep- arate. In some instances this arrangement has been found to stimulate care in picking and to accelerate markedly the rate of picking. Each picker's work is always open to inspection. With one picker on a row the natural tendency 264 Yearbook of the Department of Agriculture, 1919. is to induce the slower pickers to keep up with the faster workers. Inasmuch as the field boxes are near the tree being picked, this arrangement does away with the necessity for each man walking with his filled picking sack from the tree to the box row, as was formerly the case, and in this way saves considerable time. Extensive experience with com- mercial individual-tree picking work during the past eight Weighing, an Essential Step. FIG. 11. Recording the weight of lemons produced by a Eureka tree at the time of one of the regular monthly pickings. The arrangement of truck and scales is convenient for securing the weight of fruit where this method of keep- ing individual-tree records is desired. years .has shown that the cost of picking the crops in this way is not much, if any, greater than where the crops are picked in the ordinary manner. RECORDING INDIVIDUAL-TREE PERFORMANCE. The foreman of the picking crew usually records the individual-tree yields, as shown in figure 11. Each day, after the trees have been picked and before the boxes of fruit are assembled for transporting to the packing house, Cooperative Improvement of Citi^us Varieties. 265 or at convenient times during the day, the foreman records in a field notebook the number of boxes picked from each tree. The partly filled boxes are usually recorded as esti- mated fractional parts of a full box, frequently as eighths. Some growers do not consider this estimate accurate enough and weigh each partly filled box and record its weight of fruits in terms of pounds and ounces. A convenient and widely used form for recording the yield of each individual tree is as follows : Variety Date Tree No. Boxes. Part boxes. Quality. Notes. j 2 3 - If more than one picking is made from each tree, enough additional columns are provided to care for these data. Where this form is used it is only necessary for the fore- man to insert the name of the variety, the date of picking, and the block and row number on each page. Care is taken to look at the tree number each time before recording the data, in order to be sure that no mistake is made. In addition to the number of boxes of fruit borne by each tree the foreman usually makes a note of the apparent quality of the fruits and of any unusual tree condition. These notes are usually made by means of symbols, as, for example, A, for first grade; B, for second grade; and C, for culls. A tree showing evidences of disease is marked by recording X along with the yield data. Various amplifi- cations of this system are in use in many orchards, and have been found to be of great service in giving the trees individ- ual attention and care. COOPERATION IN SECURING AND DISTRIBUTING BUD WOOD. The California Fruit Growers' Exchange, a cooperative organization of about 10,000 members, recognizing the com- 266 Yearbook of the Department of Agriculture, 1919. mercial importance of this work, established in May, 1917, a department of bud selection. The work of this depart- ment is to secure bud wood from superior performance- record trees and distribute it to propagators. The head of this department is a scientifically trained man, who is familiar with the research which has led up to the intro- duction of improved methods of securing and propagating reliable citrus bud wood. The object of the work is to put into practice the results of the investigation of this subject by the Bureau of Plant Industry in order to improve the quantity and quality of the citrus production in the State as a whole. It is looked upon by those interested as a public service, both to the producer and to the consumer, and for this reason has the whole-hearted cooperation and support of everyone concerned. This service is performed at cost, and from the beginning has been self-supporting. The operation of this department is briefly outlined in the following paragraphs. THE SELECTION OF SUPERIOR PARENT TREES. For several years preceding the establishment of the bud- selection department many of the leading citrus growers possessing the best orchards in the State had been keeping individual-tree records of all the trees in their orchards. Some of the largest orchards are approximately 1,500 acres in extent. The tree records of all of these orchards were made available for the work of securing and distributing reliable bud wood. A careful survey was made of these orchards, which are located in every important citrus dis- trict in California, and a detailed analysis was made of the individual-tree records of production. The orchards show- ing the best and most consistent records for each variety and those where the fruit was found to bring the highest market price in its class were selected for more detailed study. Usu- ally three or more years of individual-record keeping were required before any selection of parent trees was made. In the orchards where the conditions were found to be satisfactory for this work all the highest yielding trees were carefully inspected in connection with their past perform- ance. The type of fruit was carefully examined. The uni- Cooperative Improvement of Citrus Varieties. 267 formity of fruits on all parts of the trees was studied. All trees bearing irregular fruits or those having variable branches were immediately excluded from further considera- tion. The highest yielding trees which were found to bear uniform fruits of the best type for the variety were selected as sources of bud wood for propaga- tion. In this work the individual-tree records have been found to be invaluable. Experience has shown that an in- telligent selection of trees could not have been made without them. In addition to the records and the examination of the trees, their habit of growth, and the charac- teristics of the foliage and fruits, the selection of parent trees has been guided by an intimate knowledge of the trees of the variety gained through systematic indi- vidual-tree record work by those having a nat- ural inclination for it. KIND OF BUD WOOD. Fruit-Bearing Orange Bud Wood. FIG. 12. Typical fruit-bearing Valencia orange bud stick, showing the type of bud wood secured for propagation. Only fruit-bearing bud wood is cut from the parent trees for propa- gation. Usually only those bud sticks are secured which have one or more typical fruits attached, as shown in figure 12. As a rule, 5 large viable buds are obtained on each orange bud stick and 10 strong buds with each lemon bud stick. The buds from this young and somewhat 268 Yearbook of the Department of Agriculture, 1919. immature growth have been found, both experimentally and commercially, to give better results in propagation than the buds from older growth or from sucker wood. On the average, 500 good buds are secured from each full-bearing parent tree during a season. HANDLING THE BUD WOOD. The bud sticks from each parent tree are kept in separate bundles. A tag with a serial number is attached to each bundle. A duplicate tag with the same serial number, the number of the tree from which the buds were cut, and the name of the propagator to whom the buds are to be sent is filed in the bud-selection department. With this informa- tion, together with the individual-tree records, it is possible at any time to trace any progeny in a nursery to the parent tree and to examine the performance record of the parent tree for the information of the nurseryman, a prospective purchaser of the progeny trees, or any other interested person. The leaves of each bud stick are trimmed off immediately after cutting, as shown in figure 13. As soon as all the bud sticks desired are secured from a tree, they are tied in a bundle, tagged, and packed in moist, sterile sphagnum moss. Several bundles of bud wood are usually packed tightly together, and this package is covered with strong burlap. These packages are kept in a cool temperature, preferably about 70 F., until the bud wood is delivered to the prop- agator. Under these conditions citrus bud wood can be kept safely for several weeks. However, experience has shown that it is desirable to use the buds as soon as possi- ble after cutting them from the parent trees. COST OF THE WOOD. Inasmuch as the business of securing and distributing these bud is conducted by a cooperative nonprofit organi- zation, the buds are supplied to propagators at cost. At the present time a charge of 5 cents is made for each good bud to members of the cooperative organization or 6 cents for each bud to propagators who are not members of the organization. As soon as the volume of business warrants, Cooperative Improvement of Citrus T arieties. 269 this cost will be reduced. The owners of the trees from which the buds are cut are paid 1| cents for each bud secured from their trees. Fruit-Bearing Lemon Bud Wood. FIG. 13. Two typical bud sticks on a superior Eureka lemon parent tree. The leaves have been cut off the one on the right in order to show the method of preparing the bud sticks for packing. The cost of maintaining this bud-selection department includes the pa3 7 ment for the buds to the owners of the parent trees, the assembling, tabulating, and studying of extensive individual-tree data, the selection of the superior 270 Yearbook of the Department of Agriculture, 1919. parent trees, collecting information regularly as to the be- havior of the buds and the trees grown from them, and the survey of new orchard areas for the location of additional parent trees. In 1919 an experimental citrus nursery of 7 acres was established for the purpose of trying out different methods of budding, determining the comparative value of different kinds of stocks, and securing other important information for the benefit of the propagators and the growers. USES OF SELECTED BUDS. The buds secured from the superior parent trees are be- ing extensively used by growers for top-working undesirable or drone trees in established orchards or for top-working the trees of one citrus variety with another and by propa- gators who are growing trees for sale or for their own plant- ing. Up to this time a large proportion of the buds have been sold to nurserymen, who quickly realized the impor- tance of furnishing to planters trees grown from reliable buds. An illustration of nursery trees grown from these buds is shown in figure 14. In fact, under present conditions it is almost impossible for nurserymen in California to sell at any price any other kind of citrus trees. The trees grown from the selected buds sell for a much greater price than the added cost of the buds to the nurserymen. The increas- ing appreciation by citrus growers of the importance of planting good trees makes it seem certain that the utiliza- tion of this work will be greatly increased in the near future. In the following table the development of the bud-selec- tion service is shown by the number of buds sold each season from the inauguration of this work to date : Buds sold from superior parent trees. Year and budding season. Number of buds sold. Year and budding season. Number of buds sold. Season of 1917: Spring 25 550 Season of 1919: Fall 82 850 Fall Season of 1918: Total Fall SS 958 Cooperative Improvement of Citrus Varieties. 271 Out of the total number of buds distributed approxi- mately 75,000 were used for top-working established unde- sirable trees, and the remainder were used by propagators for propagating nursery trees. These buds were secured from superior parent trees in 21 orchards located in southern Cali- fornia. During the war comparatively little citrus propagation was carried on. Since the close of the war California nurserymen have planted more than 100 bushels of citrus seed for growing stocks. This recent great activity in stock production indicates that there will be a very largely in- creased demand for the selected buds for use in budding this stock in the near future. SECURING RELIABLE TREES. The bud-selection department maintains an office where records are kept of all the available trees for sale that were grown from the selected buds furnished by that department. The parentage of these trees, their condition of growth, and other details are furnished to all inquirers without cost. From these data the planters can intelligently decide where to buy reliable and satisfactory trees. This service is prov- ing to be an invaluable aid to citrus growers. The widespread membership of the cooperative organiza- tion, continually advised as to the progress of the work of bud selection arid propagation, has been the most effective way through which this information has been made avail- able to the citrus industry as a whole. The officials of the State University and the United States Department of Ag- riculture, farm journals, and horticultural clubs have co- operated in bringing this work to the attention of all inter- ested persons. At present there seems to be no good reason why every prospective planter in California should not be able to secure reliable information as to sources of good citrus trees for planting. RESULTS OF BUD SELECTION. Extensive orchards of all the important commercial varie- ties, in which the trees were propagated from carefully selected buds secured from superior performance-record 272 Year-look of the Department of Agriculture, 1919. trees, are now in bearing in California. Careful surveys of these orchards have shown without any doubt that they are superior to comparative orchards in which the trees were Strain Characteristics Revealed in the Nursery Trees. FIG. 14. Nursery trees of the best strain of the Eureka lemon variety two years after budding on sour-orange stock. These young trees blossomed and small fruits developed while still in the nursery row. This is characteristic of the young trees propagated by the use of the improved methods described in this article. propagated without care in bud selection. It is becoming increasingly difficult to find } T oung citrus orchards where the trees were propagated without the use of carefully se- lected buds. This demonstration of the superiority of the DIVISION OF SUBTROPICAL HOW1CWJW: t COLLEGE OF AGRICULTURE BERKELEY, CALIFORNIA Trees Produced from Selected Buds. FIG. 15a. A typical 3-year-old Eureka lemon tree in a large com- mercial orchard, showing the early production of uniformly good fruits secured from trees propagated from fruit-bearing wood buds selected from superior performance-record parent trees. FIG. 151). A 3-year-old Marsh grapefruit tree in a commercial orchard, showing the heavy production of uniformly desirable fruits developed by trees propagated from selected buds secured from superior performance- record parent trees. $$**' Cooperative Improvement of Citrus Varieties. 275 trees grown from buds secured in the manner described in this article has been the compelling force that has made the bud-selection work commercially successful. The trees grown from the selected buds have shown un- usually early production of heavy crops, as shown in figure 156, and are bearing regular crops of uniformly superior quality; in other words, they are producing fruits similar to those borne by the parent trees. This uniformly good production, an example of which is shown in figure 15a, has been achieved at no greater cost than the irregular crops having a considerable proportion of fruits of worthless strains, produced by mixed-strain trees, in the ordinary orchard. The uniform fruits on the trees grown from the selected buds reduce the cost of assorting and packing the crops, compared with the ordinary crops. The uniform market grades made possible by the uniformity of fruits increase the confidence of the consumer in the fruit and induce a larger consumption. This condition is econom- ically valuable, both to the producer and to the consumer; it stabilizes the industry as a whole and adds materially to the reputation and value of the crops. COOPERATION AN ESSENTIAL. The utilization of the results of scientific research in the improvement of citrus fruits through bud selection has largely been made possible through an organized citrus in- dustry. "While the investigation of this subject could proba- bly have been carried on" without this organization, it was as a matter of fact largely encouraged and fostered by it. In the opinion of the writer the widespread use of the im- proved methods of bud selection and propagation could not have been so quickly and efficiently introduced commer- cially in the citrus industry without the active participa- tion of the cooperative growers' organization, the California Fruit Growers' Exchange. O 210105 UNIVERSITY OF CALIFORNIA LIBRARY Los Angeles This book is DUE on the last date stamped below. Form L9-75m-7,'61(Cl437s4)444 LOS AN GJBLB8 LIBRARY 3 boo' on the last date stamue, SB 369 A 001 095 390