UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA PRUNING AND THINNING EXPERIMENTS WITH GRAPES A. J. WINKLER BULLETIN 519 November, 1931 UNIVERSITY OF CALIFORNIA PRINTING OFFICE BERKELEY, CALIFORNIA CONTENTS PAGE The influence of pruning and crop on vine capacity for growth and fruiting 3 Object of the investigation 4 Plan of the investigation 4 Definition of types of pruning used 5 The effect of less pruning on vine growth 7 Increase of leaf activity 7 Earlier development of shoots 8 Increase of bulk of the vine 10 The effect of crop on vine growth 10 The effect of less pruning on fruiting 12 Increase of capacity for bearing 12 Improvement of quality 13 Development of flowers 13 Decrease in coulure (shelling) and the setting of shot berries 13 Increase in size of cluster parts 15 Discussion and application 16 The utilization of longer pruning ., 19 Crop regulated entirely by thinning 20 Crop regulated entirely by pruning 20 Crop regulated by longer pruning and thinning 21 Application to practice of methods of longer pruning with thinning to regulate crop 21 Definition of types of thinning and pruning used 22 The responses of varieties subject to coulure, setting of shot berries, or both 24 Muscat (of Alexandria) 24 Dattier (Dattier de Beirut) 29 Hunisa 30 Molinera (Red Malaga) 33 The responses of varieties that usually set well-filled clusters of normal berries 36 Malaga 36 Emperor 39 Ohanez (Almeria) 41 Tokay 43 Discussion and recommendations 46 Plower-cluster thinning 47 Cluster thinning 48 Berry thinning 49 Recommendation for longer pruning with thinning 51 Other varieties 54 Cost of longer pruning and thinning 55 Longer pruning 55 Thinning 55 Cautions 56 PRUNING AND THINNING EXPERIMENTS WITH GRAPES A. J. WINKLERi THE INFLUENCE OP PRUNING AND CROP ON THE CAPACITY OF THE VINE FOR GROWTH AND FRUITING The literature of vine pruning, although vast in amount, contains but few references to the influence of pruning" on vine growth where the influence of crop was eliminated. The pruning investigations of the past have dealt principally with the art of pruning. Hence, the practical art of pruning became well established centuries ago. With only minor changes, such as the location of the fruiting units, the pruning of'vinifera grapes today is the same as it has been for several centuries past. In the early investigations the effects of pruning on vine growth and fruiting were not separated from the accompanying influence of crop. Hence there was little progress in the formulation of sound theories concerning the physiological responses of the vine to pruning. In fact, the failure to separate the effects on the growth of the vine due to pruning from those due to crop have led to the acceptance of the belief that dormant pruning invigorates the vine — the more severe the pruning the greater the invigoration. It was not appreciated that the increase in crop accompanying the longer pruning, rather than the longer pruning itself, was responsible for depressing vine growth. Or, on the contrary, that the more vigorous shoot growth on severely pruned vines was largely the result of a reduction in crop and only partially owing to the restriction of the number of shoots that developed as a result of the severe pruning. This misconception concerning the response of the vine to pruning, i.e., its supposed invigorating effect was due also in part to a loose use of the term 'vigor.' The term 'vigor' was used both qualitatively in the sense of 'activity,' or rate of growth and quantitatively in the sense of ' capacity for action. ' A young vine may show great vigor in the qualitative sense of rate of growth and yet its vigor in the quanti- tative sense of 'capacity for action,' i.e., for growth and production, i Asseciate Professor of Viticulture and Associate Viticulturist in the Experi- ment Station. 4 University of California — Experiment Station may be much less than that of an. old and relatively inactive vine. Similarly, if a vine is pruned severely, the number of shoots the vine produces is reduced, and the shoots may be more vigorous than the shoots of a lightly pruned vine. The vine will appear more vigorous, but in the quantitative sense it has less capacity for production ; that is, it will make less total growth. On a single shoot, on the other hand, vigor and capacity for action vary together ; that is, a vigorous shoot has large capacity and a weak shoot small capacity. To aid in a better understanding of the characteristic responses of the vine to pruning, vigor and capacity will be denned as follows: Vigor is the quality or condition of active growth of the vine or a part of the vine. Capacity is the quantity of action in respect to growth and produc- tion of which the vine or a part of the vine is capable. Although the influence of pruning on vigor is utilized in develop- ing the desired form of trunk in the training of young vines, once the vineyard is established the grower is primarily concerned with the capacity of the vines to produce large crops of good fruit over a long period of years. It must be borne in mind, however; that the capacity to produce fruit is dependent on the production of wood; hence, to produce heavy crops over a period of years a vine must not only be capable of maturing a good crop of fruit in a given year but also of maturing a good growth of wood. Objects of the Investigation. — In the hope of obtaining more defi- nite information concerning the responses of the vine to dormant pruning, an investigation was started at Davis in 1921. The specific objects of this investigation were to determine (1) the effect of dor- mant pruning on vine growth, (2) the effect of crop on vine growth, and (3) the effect of dormant pruning on fruiting. After the discovery of the beneficial effects of decreased pruning on growth and fruiting, the investigation was broadened to determine the possibility of retaining these benefits with types of pruning which are less severe than most of those now generally employed, but which are still commercially practicable. PLAN OF THE INVESTIGATION The plantings for the investigation comprised at first three varie- ties — Muscat (of Alexandria), Black Monukka, and Alicante Bouschet. With Muscat and Black Monukka the pruning treatments extend along the rows, and these rows were broken up into four parts by the two different distances of planting. In the two parts planted Bul. 519] Pruning and Thinning Experiments with Grapes 5 6 x 12 feet there were 7 vines of Black Monukka and 8 vines of Muscat of Alexandria under each type of pruning" and in the two parts planted 12 x 12 feet there were 4 vines of each variety under each type of pruning. The Alicante Bouschet plantings were 12 x 12 feet. They were divided into two plots with 8 vines under each type of pruning in each plot. The soil in which the vines were planted was a deep and uniform Yolo fine sandy loam. The land was leveled before planting to facili- tate irrigation. DEFINITION OF TYPES OF PRUNING USED 1. Nonpruned, no crop: no pruning; all clusters removed approxi- mately four weeks before blooming. la. Nonpruned, part crop: no pruning; all clusters in excess of what was thought necessary for a good crop of fruit were removed approximately four weeks before blooming. lb. Nonpruned, all crop: no pruning; all clusters allowed to develop. Figure 1A shows a nonpruned vine which is typical of the form of vines in pruning types 1, la, and lb. 2. Cane pruning, part crop: pruning similar to the commercial practices of cane pruning as used on Thompson Seedless (Sultanina), except that more wood (four to seven canes 2 to 4 feet long) was retained (fig. IB). All clusters in excess of what was thought neces- sary for a good crop of fruit were removed approximately four weeks before blooming. 3. Normally (spur) pruned, no crop: pruning as nearly as possible in accord with the best-accepted commercial practices of pruning these varieties; 2 all clusters removed before blooming. 3a. Normally (spur) pruned, all crop: pruned the same as the normally pruned, no crop (3) ; all clusters allowed to develop. Figure 10 shows a normally pruned vine which is typical of the form of vines in pruning types 3 and 3a. 4. Severely pruned, no crop: pruning similar to that of the nor- mally pruned vines, but more severe, only the base buds being re- tained on the spurs ; all clusters removed before blooming. 4a. Severely pruned, all crop: pruned the same as the severely pruned, no crop (4) ; all clusters allowed to develop. Figure ID shows a severely pruned vine which is typical of the form of vines in pruning types 4 and 4&. 2 Bioletti, F. T., and H. E. Jacob. Head, cane, and cordon pruning of vines. California Agr. Exp. Sta, Oir. 277:1-32. 1924. University of California — Experiment Station Fig. L- Vines Bhowing the different types of pruning used: A, nonpruned; B, cano pruned; C, normally pruned; 1), severely pruned. Bui,. 519] Pruning and Thinning Experiments with Grapes THE EFFECT OF LESS PRUNING ON VINE GROWTH The principal effect of less or no pruning during- the dormant season which influences vine growth is an increase in leaf activity. Accompanying the increased leaf activity there is an earlier develop- ment of the shoots and an increase of the bulk of the whole vine. Increase of Leaf Activity. — The total leaf activity of the vine depends on the number, size, and quality of its leaves and on the time during which these leaves are active. In determining the influ- ence of pruning on the number and weight of vine leaves, counts were made on five vines each of Muscat and Alicante Bouschet under each type of pruning on May 5, June 24, and October 15 during one season. In arriving at the weight of leaves to a vine at the time of counting, 1,000 representative leaves were removed from other vines under each type of pruning for weighing. Table 1 shows that the smaller the amount of wood removed by pruning, the more rapidly the leaves develop and the greater the total foliage produced during the next growing season. TABLE 1 The Weight and Estimated Activity of the Leaves with the Several Types of Pruning on Muscat and Alicante Bouschet Type of pruning No pruning, part crop...... Cane pruning, part crop. Normal pruning, all crop Severe pruning, all crop. Weight of leaves May 5 June 24 October 15 pounds pounds pounds 8.5 13.1 19.4 1.9 7.4 14 9 0.4 3 3 7.6 01 1.8 6.7 Relative weight; no pruning = 100 October 15 weighing Average of the three weighings 100 52 24 17 As shown in table 1, the nonpruned vines had produced a greater weight of leaves by May 5 than the normally pruned vines produced by October 15. Likewise, the nonpruned vines had a greater weight of leaves to a vine on May 5 than the cane-pruned vines had on June 24. In other words, it required only 35 days for the nonpruned vines to produce 8.5 pounds of leaves, while 85 days were required for the cane-pruned vines to produce 7.4 pounds of leaves, and 198 days for the normally pruned vines to produce 7.6 pounds of leaves. The great advantage in total weight of leaves to a vine in favor of the unpruned vines prevailed throughout the growing season, and on October 15 they had 1.3 times the weight of leaves of the cane-pruned 8 University of California — Experiment Station vines, 2.5 times that of the normally pruned vines, and 3.0 times that of the severely pruned vines. The above influence of pruning on leaf development is substantiated by a large number of leaf weigh- ings for the varieties used in the tests discussed in the second part of this paper. The effect of type of pruning on leaf activity is indicated in the second column from the right of table 1. If total weight (which is reduced to relative weights to facilitate comparisons) on October 15 alone is considered cane pruning reduced leaf weight 23 per cent, normal pruning 60 per cent, and severe pruning 65 per cent. This comparison, however, does not take into account the earlier develop- ment of the leaves and the consequently longer period of time during which the leaves on the vines with decreased pruning function. A comparison which makes use of both weight and time would give a more accurate indication of the retarding effect of pruning on leaf activity. To make such a comparison the ratios of the weights of the leaves for the several types of pruning at each date of weighing to that of the nonpruned vines was obtained. The average ratios for the three weighings are shown in the column at the extreme right of table 1. Here again the average of the weights at the several dates of weighing for the nonpruned vines is listed as 100. By using the average ratios for the three weighings under each type of pruning, the loss of leaf activity by cane pruning is 48 per cent, that by normal pruning is 76 per cent, and that by severe pruning is 83 per cent. The percentage reductions in leaf activity following the different types of pruning appear large, but they are corroborated by measurements of the effects of pruning on the increase of bulk of the whole vine and on the character of the fruit and the amount of crop borne. Earlier Development of Shoots. — The graphs of figure 2 indicate that the type of pruning modifies both the rate of shoot growth and the amount of total length growth. The effect on rate of length growth is very similar to that of leaf development shown in table 1. The importance of this effect is seen when the amount of length growth found at the several dates of measurement is taken into account. On May 13 the nonpruned vines had made 74 per cent of the total season's length growth, while the normally and severely pruned vines had made only 23 and 12 per cent respectively. Furthermore, at this time the nonpruned vines had made 90 and 95 per cent respectively as much length growth as the normally and severely pruned vines made during the entire season. The graphs of figure 2 indicate a very rapid elongation of shoots for the nonpruned vines during April and early May. Owing to the Bul. 519] Pruning and Thinning Experiments with Grapes 9 many shoots that develop, however, the rate of growth rapidly dimin- ishes so that very little additional elongation occurs between July 1 and the end of the growing season. With this type of growth there is very little competition between additional vegetative growth and the fruit for the food materials in the vine after midseason when the fruit is maturing. The more serevely pruned vines, on the contrary, start growth later, and, owing to the relatively few shoots that develop, the rapid growth characteristic of the first part of the season in all vines continues to the end of the growing season. Thus the competition for the food material in the vine by the fruit and new vegetative growth n ON-PRUNED ~zz-~? CANE PRUNED —-^ f f" m *•* *** y /. — - NORMALLY PH UNED - r s 1 / / / / / / r' J / / / s' IV PRUNED / / / I (**£■' Fig. 2. — Pruning retards the rate and reduces the total length of cane growth. continues to be keen throughout the season. The growth curves of the cane-pruned and normally pruned vines are intermediate between that of the nonpruned and that of the severely pruned vines. It has also been shown 3 that the earlier shoot elongation of the nonpruned vines in spring does not cause so great a drain on the stored food reserves (sugars and starch) of the vine as do the type of growth made by the normally and severely pruned vines. The deple- tion is less, because of the larger storage of reserves in the vines of decreased or no pruning, and because of the greater ratio of weight of leaves to weight of new shoots in these vines during this season than 3 Winkler, A. J. The effect of dormant pruning on the carbohydrate meta- bolism of Vitis vinifera, Hilgardia 4:153-173. 1929. 10 University of California — Experiment Station in the normally and severely pruned vines. As a result of the lesser drain on the stored reserves during early summer, as well as the greater ratio of leaves to shoots, the flower clusters on the vines with decreased or no pruning are better nourished. The early moderation of length growth in these vines as indicated above also makes for a better condition of nutrition during the development and ripening of the berries. Increase of the Bulk of the Vine. — As might be expected, the growth of the whole vine varied with the weight of leaves and the time during which they were functioning. Vines which had been pruned to different degrees of severity for three, five, and seven years had the relative differences in weight shown in table 2. TABLE 2 The Kelative Decrease in Weight or Vine with Increased Pruning Type of pruning No pruning, no crop Normal pruning, no crop Severe pruning, no crop.. Alicante Bouschet 3 years old 5 years old 7 years old Black Monukka Relative weight; no pruning =100 100 75 69 The figures of table 2 show that the relative weight of vine varied, as did that of the leaves, inversely as the severity of the pruning. Since these vines were treated alike except in the manner of pruning and since bearing was prevented by the removal of all flower clusters sev- eral weeks before blooming, the differences in growth are undoubtedly due to differences in pruning. It seems logical, therefore, to conclude that, other things being equal, the less the pruning the greater the leaf activity and consequently the greater the total growth of the vine. THE EFFECT OF CROP ON VINE GROWTH It is generally recognized by grape growers that increase of crop usually follows longer or decreased pruning. 4 The effect of crop, as such, on vine growth, however, has not been clearly understood owing primarily to the fact that under vineyard conditions it is impracti- cable to separate the effects of crop and pruning. To bring out the 4 Bioletti, F. T. Vine pruning in California. California Agr. Exp. Sta. Bui. 241 and 246:29-31. 1921. Bui,. 519] Pruning and Thinning Experiments with Grapes 11 effect of crop on growth several series of vines were treated alike in every respect except crop. In one series of vines (no crop) bearing was pre- vented entirely by the removal of flower clusters several weeks prior to blooming; in another series (part crop) the crop was restricted by flower-cluster removals to what was thought to be a good crop for the vine; while in a third series (all crop) all of the clusters that were formed were allowed to develop to maturity. Table 3 shows the mean decrease in vine weight due to increases in crop. TABLE 3 The Effect of Bearing on the Growth of Vines as Indicated by Total Weight Black Monukka Alicante Bouschet Type of pruning Average weight of crop (last 4 crops) Average total weight of vine (vines 7 years old) Average weight of crop (4 crops) Average total weight of vine (vines 5 years old) No pruning, no crop pounds 28 53 22 pounds 61 49 39 48 42 pounds 36 42 22 pounds 63 35 No pruning, all crop 25 38 Normal pruning, all crop 21 The figures of table 3 show that within the limits of this experi- ment there is a regular falling off in the weight of the vines with an increase in crop. The depressing effect of bearing on growth, as indicated by total weight of vine, however, is not the same with the different types of pruning. An average annual crop of 36 pounds during four years on the nonpruned, part-crop Alicante Bouschet vines reduced their weight only 44 per cent below that of the nonpruned, no-crop vines, while an average annual crop of 22 pounds during four years reduced the weight of the normally pruned, all-crop vines 66 per cent below that of the nonpruned, no-crop vines. In the case of the normally pruned, all-crop vines this reduction was only in part due to crop since these vines also had their capacity reduced by pruning. Although of smaller magnitude, the results obtained with Black Monukka are in very close agreement with those of Alicante Bouschet. These figures indicate that vines which have their capacity for growth reduced by pruning (see table 2) are more sensitive to further reduction in growth by bearing than are nonpruned vines. The depressing effect of bearing on growth increases with increased severity of pruning, or, vice versa, it decreases with decreased pruning. 12 University of California — Experiment Station THE EFFECT OF LESS PRUNING ON FRUITING The effect of pruning on fruiting is twofold. With a decrease in pruning the capacity of the vine to produce fruit is increased. In many varieties which are subject to coulure (shelling) or the setting of shot berries (miller and age) or both, the quality of the fruit is also improved by decreased pruning, provided the crop is regulated by flower-cluster thinning. Increase of Capacity for Bearing. — Table 4 shows the increase in capacity for bearing of vines with decreased pruning. TABLE 4 The Effect of Type of Pruning on the Bearing Capacity of Vines Type of pruning Average weight of crop for 4 years Relative total growth of the vines; no pruning, no crop =100 pounds 51 28 23 6 100 63 69 Normal pruning, all crop Severe pruning, all crop 65 64 Although the nonpruned vines produced an average crop of 51 pounds each year, their bulk was reduced only 2 per cent below that of the normally pruned vines which were producing average crops of 23 pounds, or only 45 per cent as much. The severely pruned vines with a very light yearly crop were limited in growth to almost the same degree as vines receiving the other types of pruning and bearing moderate and heavy crops. In the case of cane pruning the crop was restricted by flower-cluster thinning to what was thought to be its capacity, but it seems that the increase of capacity as a result of the decreased pruning was underestimated; for despite the relatively heavy average crops they produced the greatest total growth of any of the vines with crop (table 4). These figures give some indication of the limiting effect of pruning, as well as of crop, on the capacity of the vine. Crop and pruning (see also table 2) apparently diminish the capacity to about the same extent. With decreased pruning, capacity increases to such an extent that larger crops with these types of pruning are, if kept within suit- able limits, no more depressing to vine bulk (growth) than are the Bul. 519] Pruning and Thinning Experiments with Grapes 13 light crops of severely pruned vines, which have had their capacity greatly reduced by pruning. Improvement of Quality. — The principal effect of decreased or no pruning which influences the quality of the fruit of varieties which are subject to coulure and the setting of shot berries, or both, is an improvement in the development of the flower parts. As a result of this better development, there is an increase in both the number and percentage of normal berries to a cluster, and an increase in the size of the cluster parts (table 7). Development of Flowers. — Although pruning retards the develop- ment of both the pollen and the female parts of the flowers, the effect on the pollen appears to be the more pronounced. The effect of type of pruning on the germinability of pollen is shown in table 5. TABLE 5 Decrease in Germinability of Pollen with Increased Pruning; Average for Three Years for the Three Varieties Employed Type of pruning Germination per cent 44 55 46 14 10 The pollen of the severely and normally pruned vines had an aver- age germination of only 10 and 14 per cent respectively. The greater leaf activity of the nonpruned, all-crop vines improved the nutrition of the flowers sufficiently to bring the germination of the pollen up to 44 per cent. Where the ratio of leaves to flower clusters was still further increased by flower-cluster removals, the germination of the pollen was 46 per cent with cane pruning, and 55 per cent with non- pruning. Decrease in Coulure (Shelling) and the Setting of Shot Berries. — The improvement in the development of the flower parts has resulted not only in the setting of a larger number of berries to a cluster but in an increased percentage of normal berries. The effect of type of pruning on the number of berries and on the percentage of normal berries to a cluster is shown in table 6. Even with the very much larger crop of the nonpruned, all-crop vines, these vines produced clusters with 13 and 56 per cent respec- tively more berries than the normally and severely pruned vines. The percentage of normal berries to a cluster on the nonpruned, all-crop 14 University of California — Experiment Station vines was 18 and 30 per cent respectively greater than that of the normally and severely pruned vines. Still greater increases in the number and the percentage of normal berries to a cluster were obtained with the nonpruned, part-crop, and the cane-pruned, part-crop vines, where the ratio of leaves to flower clusters was greater. TABLE 6 The Decrease in the Number and the Percentage of Normal Berries to a Cluster with Increased Pruning ; Averages for Four Years for Muscat and Alicante Bouschet Type of pruning Number of berries to a cluster Percentage of normal berries 95 179 154 84 61 92 96 95 78 71 The beneficial influence of a large number of leaves to a vine on the set and development of the berries of Muscat is further shown in figure 3. As indicated at A, the clusters of the cane-pruned, part- crop vines are well filled with normal berries of uniform size. B shows that the clusters of the normally pruned vines are affected by coulure and that the berries are irregular in size, with a considerable number of shot berries on each cluster. The clusters from the defoliated 5 vines (C) are badly coulured and show an increased tendency to set shot berries, and the few normally seeded berries are undersized. As the quality of the flowers rather than their number to a cluster is influenced by the type of pruning, a more direct indication of the reduction in the tendency to coulure and to set shot berries (miller an- dage) with decreased pruning may be obtained by comparing the figures for the nonpruned part-crop, and the normally pruned all- crop vines of table 6. After flower-cluster thinning the nonpruned, part-crop vines, the number of clusters to a vine was about the same for these types of pruning. The normal pruning reduced the number of berries to a cluster 53 per cent. This reduction of the number of berries approximately indicates the increase in the proportion of the flowers which failed to set or coulured as a result of normal pruning. The percentage of normal berries was reduced to 78 by normal prun- ing and only to 96 by nonpruning. The difference in these numbers s Normally pruned vines, defoliated two weeks before blooming. Leaves arising after the berries had set were allowed to develop. Bul. 519] Pruning and Thinning Experiments with Grapes 15 (96-78) indicates the difference in the proportion of the berries which set but which remained small and seedless (miller andage) . Thus, by increasing the severity of pruning from none with part crop to normal with all crop, the loss from increased coulure was 53 per cent and that from increased millerandage 19 per cent, or a total of 72 per cent. This loss of 72 per cent as a result of coulure and the setting of shot berries with normal pruning is in very close agreement with the Fig. 3. — The effect of the number of leaves to a vine on the size of cluster and the setting and development of the berries of Muscat: A, cane pruned, flower- cluster thinned; B, normally pruned; C, normally pruned, defoliated. (From Hilgardia, Vol. 4, No. 6.) loss in leaf activity of 76 per cent with this type of pruning as indi- cated in table 1. The very close correlation of reduced leaf activity with the loss through coulure and the setting of shot berries further emphasizes the importance of the leaves in the production of high quality in the fruit. Increase in Size of Cluster Parts. — The effect of type of pruning on size, as indicated by length of cluster, is shown in table 7. The clusters of Muscat and Alicante Bouschet from the normally pruned all-crop vines were 8 and 2 per cent shorter respectively, and those from the severely pruned all-crop vines were 24 and 13 per cent shorter, than the clusters from the nonpruned all-crop vines. With 16 University of California — Experiment Station these varieties the increase in the ratio of leaves to cluster by flower- cluster removals before blooming on the nonpruned part-crop and cane-pruned part-crop vines has resulted in the production of very large clusters. TABLE 7 The Relation - of Length of Cluster to Type of Pruning with Muscat and Alicante Bouschet; Averages for Four Years Length of cluster Type of pruning Muscat Alicante Bouschet inches 8.5 10.8 10.5 7.8 6.5 inches 6 1 8.6 8 3 6.0 5.3 DISCUSSION AND APPLICATION Figure 4 and table 8 bring the responses of the vine to pruning and crop, as shown in the foregoing tables and figures, side by side, which helps to clarify the interrelations of these factors to one another. The depressing effect of pruning on growth is illustrated by B, D, and F, of figure 4. When the vines with no pruning (fig. 4 F) are used as a standard, normal (fig. 4 D) and severe (fig. 4 B) pruning depressed total growth 25 and 31 per cent respectively. This depres- sion represents the loss in capacity as a result of pruning. The depressing effect of crop on growth is illustrated by F, G, and H of figure 4 for the nonpruned vines. For these vines growth has been depressed 22 per cent by the part-crop treatment (fig. 4 G) and 35 per cent by the all-crop treatment (fig. 4 II). This depression of growth represents the loss in capacity as a result of crop production. Thus in these experiments pruning and crop had about the same relative depressing influence on capacity for growth. Maximum prun- ing with minimum crop (fig. 4 A) depressed growth 36 per cent and minimum pruning with maximum crop (fig. 4 II ) depressed growth 35 per cent. As might then be expected, nonpruned vines have a greater capacity for production of fruit than pruned vines. With the former there is only the depressing effect of crop, while with the latter there is the combined depressing effect of pruning and crop. This condition Bul. 519] Pruning and Thinning Experiments with Grapes 17 of greater capacity of the vines less severely pruned may be illustrated by comparing the growth made by the with-crop vines under the different types of pruning with that of the nonpruned no-crop vines. Growth was depressed 22 per cent by a mean crop of 50 pounds on the nonpruned, part-crop vines; 30 per cent by a mean crop of 37 pounds on the cane-pruned, part-crop vines; 35 per cent by a mean •2 g o GO t-i O & § E -3 ° o I Grow/// Crop 400 300 fl a POO g « o o So O Fig. 4. — The effect of pruning on the capacity of the vine for growth and production and the effect of crop on growth ; average results for the three varieties for four years: A, severely pruned, all crop; B, severely pruned, no crop; C, nor- mally pruned, all crop ; Z>, normally pruned, no crop ; E, cane pruned, part crop ; F, nonpruned, no crop ; G, nonpruned, part crop ; H, nonpruned, all crop. crop of 24 pounds on the normally pruned all-crop vines ; and 36 per cent by a mean crop of 6 pounds on the severely pruned all-crop vines. The figures indicate that as the severity of pruning is increased the smaller is the crop that a vine can bear and still make a satisfactory growth, or vice versa, a crop of given amount is more depressing on growth as the severity of pruning is increased. Pruning not only influenced the capacity of the vines with regard to total production as indicated by figure 4, but it also affected the weight of cluster, length of cluster, the number and percentage of 18 University of California — Experiment Station normal berries, and the germinability of pollen. Since the effects of pruning on the fruiting of the several varieties were very much the same, only the results for Muscat are shown in table 8. TABLE 8 The Effect of Pruning on the Fruiting of Muscat Average weight of clusters Average length of clusters Normal berries Average Type of pruning Number to a cluster Per cent of total germinability of pollen / * S 4 5 6 pounds 0.41 1.14 1.00 0.43 0.37 inches 8.5 10.8 10.5 7.8 6.5 71 127 115 43 35 per cent 93 95 94 68 64 per cent 40 No pruning, part crop 51 42 Normal pruning, all crop 12 10 As shown in column 6 of table 8, the pollen of the severely and normally pruned vines had an average germination of only 10 and 12 per cent respectively. The greater ratio of leaves to flower clusters on the nonpruned all-crop vines improved the nutrition of the flowers sufficiently to bring the germination of the pollen up to 40 per cent. Where the ratio of leaves to flower clusters was still further increased by flower-cluster removals, the germination of the pollen was 42 per cent with cane pruning and 51 per cent with nonpruning. Column 4 of table 8 shows the effect of pruning on the set of berries. The increase in the set of berries with decreased pruning follows very closely the improvement in the germinability of the pollen. In the case of the normally pruned vines only 68 per cent of the berries were normal, while 95 per cent of those produced by the nonpruned part-crop vines and 94 per cent of those produced by the cane-pruned part-crop vines were normal. As a result of the larger number of berries that set and the greater percentage of normal berries to a cluster, the clusters of the vines with decreased or no pruning were heavier than those of the normally and severely pruned vines. The clusters of the cane-pruned and nonpruned part-crop vines were more than twice as heavy as those of the normally pruned vines (see column 2 of table 8). As might be expected for varieties such as Black Monukka, which set practically perfect clusters with normal (spur) pruning, under most conditions, the improvement in fruiting following decreased prun- ing is less than that obtained with Muscat. The germinability of the pollen of this variety was increased from about 17 per cent with normal Bul. 519] Pruning and Thinning Experiments with Grapes 19 pruning to approximately 40 per cent with cane pruning, part-crop. As a result of the better development of the flower parts more berries were set on many clusters, when the flower clusters were thinned before blooming. This may or may not be an improvement, according to the density of the clusters. At Davis such a practice tended to make the clusters too dense and too large. The greatly increased capacity of the nonpruned over that of the normally and severely pruned vines for both growth and production is the result of better nutrition. As shown in previous publications 6 the nonpruned vines not only develop a very much greater total leaf area than the normally or severely pruned vines, but also the rate of leaf formation in the former is much more rapid in spring and early summer. These responses to pruning, together with an explanation of the cause underlying them 7 leave no justification for the belief that severe pruning improves the possibilities of the vine for either in- creased growth or production. The improvement of the nonpruned vines over that of the normally and severely pruned vines in both growth and production would seem to indicate that through modification of the present systems of pruning so as to retain more wood, production may be increased. The results indicate further that these increased crops of better quality may be obtained from vines which will, as a result of the less severe pruning, remain more vigorous than normally pruned vines bearing smaller crops. The Utilization of Longer Pruning. — To arrive at what seems a logical conclusion concerning the application to practice of the prin- ciples established by the above results, each of the types of pruning except those in which all the crop was removed will be considered and the merits and demerits of each indicated. The vines with maximum (severe) pruning (fig. 4^1) gave the smallest crop — 74 per cent less than the normally pruned (fig. 4 C) — and made the least vine growth. The fruit produced by these vines was only of fair quality. It was high in sugar content, but the clusters were small and the percentage of abnormal berries was large. At the other extreme, nonpruned all-crop vines (fig. 4 H) gave the largest crop, 170 per cent more than the normally pruned (fig. 4C). 6 Winkler, A. J. Some responses of Vitis vinifera to pruning. Hilgardia 1:525-543. 1926. Winkler, A. J. The influence of pruning on the germinability of pollen and the set of berries in Vitis vinifera. Hilgardia 2:107-124. 1926. Winkler, A. J. Improving the fruiting of Muscat of Alexandria by less severe pruning. Proc. Amer. Soc. for Hort. Sci. 24:157-163. 1928. 7 Winkler, A. J. The effect of dormant pruning on the carbohydrate meta- bolism of Vitis vinifera, Hilgardia 4:153-173. 1929. 20 University of California — Experiment Station The fruit was of inferior quality owing to its lower sugar content. The clusters were somewhat larger than those of the severely pruned vines and the berries were more uniform. Obviously, neither of these two methods is of use in practice. In the former, too much pruning has reduced the crop to such propor- tions that it is unprofitable. Too little pruning in the latter has resulted in very large crops but the fruit is of inferior quality. Between the two extremes represented by the above treatments, there are three intermediate or compromise treatments, Certain of these treatments indicate a balance of pruning, growth, and crop which favors maximum production of high-quality fruit, and other treatments, balances which, although not quite so favorable from the standpoint of yield and quality, are more adaptable to vineyard practice. Crop Regulated Entirely by Thinning. — The responses of the non- pruning part-crop vines (fig. 4 G) will be considered first. Under this treatment crop is controlled entirely by the removal of excess flower clusters before blooming. These vines have produced 105 per cent more crop than the normally pruned vines (fig. 4 C). The fruit also has been of superior quality (table 8). Accompanying the larger crops of superior fruit these vines have also made 23 per cent more growth than the normally pruned all-crop vines (fig. 4 0), and 5 per cent more growth than the normally pruned no-crop vines (fig. 4D). This compromise by thinning has resulted in large crops of superior fruit accompanied by a larger growth, which indicates a most favor- able balance of pruning, crop, and growth. The treatment, however, is impracticable, owing partially to the difficulties of support, but primarily to the high cost of the removal of excess clusters by thinning. Crop Regulated Entirely by Pruning. — Another compromise is that by pruning, which is represented by the normally pruned all-crop vines (fig. 4C). It represents a heritage of the past. Under this treatment, crop is controlled entirely by pruning. The fruit is of good quality but not equal to that produced by the nonpruned part-crop vines, and the yields are much smaller. For the varieties, such as Muscat, which are subject to couture and the setting of shot berries the fruit may be of poor quality. In this treatment, pruning, crop, and growth are balanced only at the cost of a great reduction in capacity. In view of the results here reported, this type of pruning does not seem to utilize the capacity of the vine efficiently. It is, however, an economical means, whereby fair crops of average-quality grapes can be produced with most varieties. It should continue to be Bul. 519] Pruning and Thinning Experiments with Grapes 21 the usual practice where cost of production rather than appearance and quality of the fruit is the determining factor of profit or loss. Crop Regulated by Longer Pruning and Thinning. — The third compromise is that of moderate pruning 1 accompanied by flower- cluster or other types of thinning to regulate crop (fig. 4 2£). Under this treatment sufficient wood is removed each year to maintain a convenient form of the vine and to bring the matters of support and cultural operations within the limits of economical practice. Although the pruning used in this treatment removes much fruit wood, the ultimate regulation of crop depends on the removal of excess crop by flower-cluster thinning before blooming or by cluster or berry thinning after blooming. The pruning puts a limit on the maximum number of clusters that may develop and thereby serves as a means of reducing the cost of thinning. In the tests this type of pruning with flower-cluster thinning to regulate crop resulted in crops that were 54 per cent larger than those of the normally pruned vines. The fruit also has been better than that from the normally pruned vines in sugar content, coloring, and the uniformity of size of clusters and berries. In fact, the quality of the grapes under this treatment has been equal, except for slightly smaller clusters, to that of the non- pruned part-crop vines (fig. 4 6r). Again, accompanying the larger crop (54 per cent) of superior grapes these vines have made 6 per cent more vine growth than the normally pruned vines (fig. 4 0). This compromise, as a result of its greater production of both crop and growth and its tendency to improve quality greatly, offers unusual opportunity for improving the fruiting of several table-grape varieties where perfection of cluster, large size of berry and cluster, and general appearance count for so much. APPLICATION TO PRACTICE OF METHODS OF LONGER PRUNING WITH THINNING TO REGULATE CROP In view of the marked improvement obtained in capacity for both fruiting and growth and of the improvement in the quality of the fruit, especially of the Muscat, resulting from longer or no pruning accompanied by flower-cluster thinning to regulate crop, the investi- gation was expanded to test the response to several types of longer pruning and thinning of a number of table-grape varieties. In addition to flower-cluster thinning several weeks before blooming as a means of regulating crop, cluster thinning and berry thinning after the blooming period were employed with some varieties. 22 University of California — Experiment Station DEFINITION OF TYPES OF THINNING AND PRUNING USED Flower-cluster thinning is the operation of removing flower clusters between the time the clusters come out or unfold (' cluster forms') as growth starts in spring and the time of blooming (fig. 10). Entire flower clusters are removed. In this investigation the flower-cluster thinning was carried out approximately four weeks before blooming. Cluster thinning is the operation of removing entire clusters after blooming is over and the berries have set. It is carried out as soon as the usual drop of abortive berries after blooming is complete (fig. 11). Berry thinning as used in California is the operation of removing certain parts of clusters. It is done by cutting off several branches of the cluster and usually the end of the main stem (rachis) or simply by cutting off a sufficient part of the main stem to retain only the desired number of berries (figs. 12, 13, and 14). The removal of single berries is too costly and also unnecessary in field practice. Normal (spur) pruning: pruning as nearly as possible in accord with the best-accepted commercial practices of pruning the varieties used. Since crop is regulated entirely by pruning all clusters that set were allowed to develop to maturity (figs. 1 C or 5 A). Long-spur pruning, part crop: pruning similar to that of the nor- mally pruned vines, but less severe, one, two or three more buds to a spur being retained. The amount of crop was regulated by flower- cluster thinning before blooming or by cluster or berry thinning after blooming (fig. 5 B) . Half -long pruning, part-crop: pruning similar to the above, but still less severe, five to eight buds being retained on the spurs (fig. 5(7) . Provision was also made for the production of suitably placed wood for the following year by leaving renewal spurs. The amount of crop was regulated by flower-cluster thinning before blooming or by cluster or berry thinning after blooming. Cane pruning, part-crop: pruning similar to the commercial prac- tices of cane pruning (fig. IB) as used on Thompson Seedless (Sul- tanina). The amount of crop was regulated by flower-cluster thinning before blooming or by cluster or berry thinning after blooming. Bul. 519] Pruning and Thinning Experiments with Grapes 23 4 \ / / \ ... ■ 31 \t ?^n>*£ Fig. 5. — Vines showing some of the different types of pruning: A, normally pruned; B, long-spur pruned; C, half -long pruned. 24 University of California — Experiment Station THE RESPONSES OF VARIETIES SUBJECT TO COULURE, SETTING OF SHOT BERRIES, OR BOTH In respect to their responses to longer pruning and thinning, grape varieties may be divided into two groups. One group is made up of the varieties such as the Muscat, which are subject to coulure (shelling) or to the setting of shot berries or both, with normal prun- ing. The other group comprises those varieties, such as the Malaga, which usually set well-filled clusters of normal berries with normal pruning. There is much variation in the susceptibility of different varieties to coulure or the setting of shot berries, or both. One variety may be subject to both of these defects, another variety to only one. The degree to which varieties are subject to one or both of these defects also varies. This being the case, it might be expected that the response of different varieties to similar types of pruning and thinning would not be the same. In order, therefore, to give as accurate an indication of the response of each of the varieties tested to longer pruning and to give a better indication of the possibilities of longer pruning with thinning as a means of improving quality, each variety will be dis- cussed by itself. Muscat (of Alexandria). — The experiments to test the practica- bility of longer pruning for Muscat were started in 1925. One lot of 140 four-year-old vines was cane pruned, another lot of 56 vines was half-long pruned, and a third lot of 140 vines was normally pruned. On three-fourths of the cane-pruned and on one-half of the half -long pruned vines all clusters in excess of what was thought necessary for a good crop of fruit were removed approximately four weeks before blooming. The amount of crop on one-fourth of the cane-pruned vines was regulated by cluster thinning after blooming. Beginning in 1927 a series of 35 Muscat vines was pruned to long spurs. The crop of these vines was regulated by flower-cluster thin- ning. During 1929 and 1930 the crop on 35 normally pruned vines was restricted to one cluster on a shoot by flower-cluster thinning. The influence of these different types of decreased pruning and the regulation of crop by flower-cluster and cluster thinning are shown in table 9. As indicated by the figures in table 9, there was an improvement in the fruiting of the Muscat with each type of longer pruning when the crop was regulated by flower-cluster thinning. The improvement Bul. 519] Pruning and Thinning Experiments with Grapes 25 was inversely correlated with the severity of pruning; that is, the greater the amount of wood retained at pruning the greater the im- provement in fruiting. If the normally pruned vines are taken as standard, flower-cluster thinning increased the weight of cluster 58 per cent with normal pruning, 72 per cent with long-spur pruning, 89 per cent with half-long pruning, and 133 per cent with cane pruning. These increases in weight of cluster resulted from the setting of more and better berries on the vines with longer pruning. TABLE 9 The Influence of Different Degrees of Longer Pruning with Flower-Cluster, Cluster, or No Thinning on the Fruiting of Muscat Type of thinning Crop to a vine Average weight of cluster Average length of cluster Normal berries Type of pruning Number to a cluster Per cent of total pounds 30.0^ 24.5' 28.4 29.7 33.5 32.2 31.1 38.7 pounds 0.43 0.68 0.74 0.85 0.35 1.00 0.46 0.38 inches 7.8 8.6 9 5 100 7.6 105 7.8 7.7 43 65 84 94 48 115 44 49 per cent 68 Normal (spurj* Long spurt Flower cluster; 1 cluster per shoot Flower cluster; 3 clusters to 4 or 82 90 Half-long* Flower cluster; about 2 clusters to 3 shoots 93 Half-long* Cane* None 76 Flower cluster; 1 cluster to 2 94 Cane* Cluster 74 Cane* 75 * Averages for five years, t Averages for two years. The clusters of the normally pruned vines had only 43 normal berries, while those of the cane-pruned vines had 115, those of the half-long pruned vines 94, those of the long-spur pruned vines 84, and those of the normally pruned flower-cluster thinned vines 65 normal berries. That the berries of the vines of ]onger pruning were of better quality is indicated by the percentage of berries per cluster that were normal (table 9). The greater improvement in the quality of the fruit of the cane- pruned vines over that of the half-long or long-spur pruned is in strict accordance with the improvement over normal pruning of each of these other types of longer pruning. That is, a greater ratio of leaves to retained flower clusters was obtained with cane pruning than with half-long or long-spur pruning. Similarly, the ratio of leaves to retained flower clusters with half-long pruning was greater than with long-spur pruning. 26 University of California — Experiment Station The beneficial effect of an increased ratio of leaves to flower clusters is further substantiated by the illustrations of figures 6, 7, and 8. Fig- ures 6 and 7 show the influence of the increase in ratio of leaves to Fig. 6. — The crop of a representative Muscat vine that was normally pruned. Fig. 7. — The crop of a representative Muscat vine that was cane pruned and on which the crop was regulated by flower-cluster thinning. flower clusters on fruiting as obtained by cane pruning with flower- cluster thinning to regulate crop as compared to normal pruning. Figure 8 shows the additional improvement in fruiting resulting from a further increase in the ratio of leaves to flowers which was obtained Bitl. 519] Pruning and Thinning Experiments with Grapes 27 by first flower-cluster thinning, as in figure 7, and then pinching off the apical one-third of the retained clusters several weeks prior to blooming. Longer pruning without thinning to restrict crop, on the other hand, has resulted in the production of fruit of poorer quality than that of the normally pruned vines. With both the cane and half-long pruning without thinning the amount of crop to a vine was so much increased that the clusters were shorter and their weight was less, even though the number of berries to a cluster was slightly larger than that of the normally pruned vines. The clusters also were very Fig. 8. — The crop of a representative Muscat vine that was cane pruned and flower-cluster thinned to regulate crop and on which the apical one-third of the retained flower clusters was pinched off at the time of thinning. straggly. The lighter clusters with more berries to a cluster than those of the normally pruned vines show the injurious effect of this treatment on berry development. The sugar content, as shown in table 11, was low, never getting above 20° Balling. The fruit of the cane-pruned vines on which the crop was con- trolled by cluster thinning after the blooming period was no better as regards weight of cluster and number of berries to a cluster than that of the normally pruned vines, but the berries were large and the sugar content was high. Owing to the large number of clusters on the vines before and during blooming, however, the flower parts were so poorly nourished that the clusters were very straggly. Improvements in the fruiting of the Muscat similar to those ob- tained at Davis as a result of longer pruning with flower-cluster 28 University of California — Experiment Station thinning have been secured in tests with cane pruning in Madera and Tulare counties. The average results of the tests are shown in table 10. The figures of table 10 show an increase of crop over normal pruning of 71 per cent in Tulare County and of 61 per cent in Madera County in 1927. This is an actual increase in crop per acre of 5 and 3 tons, respectively. The percentage increase in crop in 1928 was larger for the Tulare plot and somewhat smaller for the Madera plot. Accompanying the large increase in crop, the weight of cluster, length of cluster, and the number and percentage of normal berries to a cluster were increased very much as a result of cane pruning with TABLE 10 The Effect of Longer Pruning with Flower-Cluster Thinning on the Fruiting of the Muscat in Madera and Tulare Counties County Type of pruning Average crop to a vine Average weight of cluster Average length of cluster Normal berries Year Number to a cluster Per cent of the total 1927 Tulare pounds 24.4 42.0 13.0 39.6 16.9 27.5 17.5 238 pounds 0.26 0.70 0.25 0.95 0.25 0.55 0.33 0.60 inches 7.4 10.8 8.6 11.2 7.1 10 3 8.2 10.1 27 62 30 90 23 58 31 53 per cent 52 Tulare \Cane 84 1928 /Normal \Cane 56 Madera 89 1927 66 Madera 82 1928 (Normal 57 75 flower-cluster thinning. Although the actual weight of cluster, length of cluster, and number and percentage of normal berries obtained with cane pruning in these tests were not so great as those obtained at Davis, the percentage increases over normal pruning compare very favorably. The smaller actual increases appear to be owing to a tendency toward poorer fruiting of all vines in these tests. The improvement in the nutrition of the flower parts before and during blooming which has resulted in the production of better fruit by the vines with longer pruning and regulation of crop by flower- cluster thinning is only one phase of the improved condition of nutri- tion of the vine following this type of pruning. In fact, the condition of nutrition of the entire vine is more favorable to the production of large crops of high-quality fruit. Responses indicating this better nourishment of the vine and vine parts with longer pruning and flower-cluster thinning are shown in table 11. Bul. 519] Pruning and Thinning Experiments with Grapes 29 As shown by the figures of table 11, the pollen of the normally pruned vines gave a germination of only 12 per cent and that of the vines of longer pruning without thinning 14 per cent as compared to 26 per cent for that of the half-long pruned vines and 42 per cent for the cane-pruned vines which were flower-cluster thinned. The better development of the flower parts brought about the better setting of the berries and subsequently increased the weight of the clusters on these vines. TABLE 11 The Influence of Longer Pruning with Flower-Cluster, Cluster, or No Thinning on the Germinability of Pollen, Sugar Content of the Fruit Juice, and the Weight of the Prunings of Muscat : Averages for Four Years Type of pruning Type of thinning Germinability of pollen Sugar content of fruit juice Annual weight of prunings None per cent 12 42 14 14 26 14 ° Balling 24.6 25.7 25.5 17.8 25 2 18.1 pounds 3.7 5.4 4 1 3.2 4 5 Half-long None 3.3 The figures of table 11 which are concerned with the increase in pruning weights and sugar content of the fruit give some indication of the better nutrition of the vine as a whole. Although the vines with cane pruning and regulation of crop by flower-cluster thinning have consistently produced larger crops than the normally pruned vines, these crops also have been higher in sugar content and the vines have made more growth. Battier {Battier de Beirut). — With the Dattier only normal and cane pruning were employed. This was owing to the limited number of vines that were available in 1927 when the tests were started. Eight vines were used with each type of pruning. The crop on all of the cane-pruned vines was regulated by flower-cluster thinning. The response of this variety to these two types of pruning is shown in table 12. As indicated by the figures of table 12, the fruiting of the Dattier was improved by longer pruning with crop regulation by flower- cluster thinning. This type of pruning resulted in an increase of 74 per cent in weight of cluster, of 22 per cent in length of cluster, of 69 per cent in the number of normal berries to a cluster, and of 18 per cent in the weight of berry over that of normal pruning. The germinability of the pollen was increased more than threefold. 30 University of California — Experiment Station In addition to these improvements in the fruit there was also an increase of 31 per cent in the yield in favor of the cane-pruned vines. Although some of the improvements in the fruiting of Dattier as a result of longer pruning with crop regulation by flower-cluster thin- ning were not so marked as in the Muscat or Hunisa, the possibilities of this type of pruning as a means of insuring more regular crop are strikingly indicated. With normal pruning where crop is regulated solely by pruning, the removal of wood to restrict the number of clusters sufficiently to prevent overbearing, results in the poor nutri- tion of the retained flower clusters up to the time of blooming. Owing to the poor nutrition of the flower clusters they are weak and sensitive to unfavorable conditions, such as cold weather, rain, or hot winds TABLE 12 A Comparison of the Fruiting of the Dattier Under Normal Pruning and Under Cane Pruning with Flower-Cluster Thinning to Regulate Crop Type Crop to a vine Average weight of cluster Average length of cluster Normal berries Average weight of 100 berries Germina- of pruning Number to a cluster Per cent of the total bility of pollen Normal pounds 16 21 pounds 0.53 0.92 inches 6.7 8.2 45 76 per cent 75 85 pounds 0.86 1.02 per cent 10 Cane 36 during blooming. Many of the berries will set imperfectly or drop and in many instances entire clusters fail to set any berries. On the contrary, with long pruning and regulation of crop by flower-cluster thinning, the nutrition of the retained clusters is sufficiently improved so that they have a better chance of setting. This influence on crop is illustrated in figure 9. Hunisa. — The Hunisa is somewhat different from the Muscat and Dattier in that it is much more subject to the setting of shot berries (miller andage) than to straggliness of cluster (coulure). In fact, the number of berries to a cluster is usually as large and often larger when most of the berries are small and seedless. As indicated by the other varieties, the cause of the setting of shot berries is the same as for coulure. This indication is further substantiated by the figures of table 13. Differential pruning tests on the Hunisa were started in 1925. Ten vines each were used with normal, half -long, and cane pruning. Owing to the limited number of vines flower-cluster thinning alone was employed as a means of regulating the crop of the vines with longer Bul. 519] Pruning and Thinning Experiments with Grapes 31 B Fig. 9. — Representative vines of Dattier: A, a normally pruned vine on which crop was regulated by pruning; B, a cane-pruned vine on which crop was regu- lated by flower-cluster thinning. 32 University of California — Experiment Station pruning. The response of the Hunisa to the different types of pruning is shown in table 13. TABLE 13 The Effect of Longer Pruning with Flower- Cluster Thinning on the Fruiting of Hunisa; Averages for Four Years Crop to a vine Clusters Normal berries Type of pruning Weight Length Number to a cluster Per cent of the total Weight of 100 berries pounds 38 41 43 pounds 0.82 1.56 1.71 inches 10.0 13.8 14.4 43 148 157 per cent 39 76 87 pounds 0.39 Half-long 0.88 Cane 0.96 The improvement in the fruiting of the Hunisa as a result of longer pruning was the greatest of any variety tested at Davis. Despite an increase of 8 and 13 per cent, respectively, in weight of crop over that of the normally pruned vines, the clusters of the half- long and cane-pruned vines with flower-cluster thinning were 90 and 110 per cent heavier. Equally as large increases resulted from longer pruning in the number and percentage of normal berries to a cluster and the average weight of berry. In addition to the production of more and better fruit than the normally pruned vines, as shown by the above figures, the longer pruned vines made more growth and ripened their fruit better. The sugar content of the fruit and the increase in the capacity of the Hunisa vines with longer pruning is shown in table 14. TABLE 14 The Effect of Longer Pruning with Thinning on the Increase in Area of the Trunk, Weight of Prunings, and Sugar Content of the Fruit Juice; Averages for Four Years Type of pruning Annual increase in cross sectional area of the trunk Annual weight of prunings Sugar content of the fruit juice square inches 0.73 0.77 0.90 pounds 4.18 4.35 4.85 ° Balling 19 Half -long 21 Cane 21 Although the increases in vine growth resulting from the longer pruning, as shown by the figures on the area of the trunk and the pruning weight of table 14, are not large, the differences are all in the same direction. Statistical treatment shows the differences to be sig- nificant. Thus, decreased pruning of this variety has enabled the Bul. 519] Pruning and Thinning Experiments with Grapes 33 vines to produce more fruit, fruit of a higher quality and better appearance, and fruit of a higher sugar content, without loss of vine growth or capacity. Molinera (Red Malaga). — Of the varieties having a tendency to coulure or the setting of shot berries included in these studies, the Molinera is probably the least affected. It usually sets medium to large clusters which are fairly well filled. However, under some con- ditions, such as those prevailing at Davis, it usually sets a rather large number of shot berries. Another defect of this variety which is possibly at least as impor- tant as the setting of shot berries, since the latter may be trimmed off if they are not too numerous, is light bearing. Under most conditions in the grape-growing regions of California the Molinera produces small and very irregular crops with normal pruning. The fact that it pro- duces large crops in very favorable seasons, however, makes the use of longer pruning without thinning to regulate crop hazardous. Such a practice would result in occasional overbearing, which would greatly weaken the vines and make the fruit of very poor quality. In the pruning tests with this variety, therefore, the effect of longer pruning with thinning to control crop on the regularity and amount of production as well as on the setting of shot berries was considered. The tests were started in 1925 with 10 vines each under normal, half- long, and cane pruning. The crops of these half -long and cane-pruned vines were regulated throughout the tests by flower-cluster thinning. In 1927 additional lots of 7 vines each were added to the lots with the above types of pruning. In this year another lot of 14 vines was pruned to long spurs with flower-cluster thinning to regulate crop. Another lot of 14 vines was cane pruned with cluster thinning to regulate crop. In discussing the results of these tests the effect of the several types of pruning on crop will be taken up first. The tendency of the Molinera to produce light and irregular crops with normal pruning together with the possibility of overcoming this defect with longer pruning accompanied by thinning to regulate crop is shown in table 15. The figures of table 15 not only show an improvement in the regu- larity of production of the Molinera, but also they indicate a great increase in the average yield. The longer pruning has resulted in the development of more flower clusters than were needed for a full crop with each of these types of pruning. Then by thinning these flower clusters judiciously a full crop for the given vines was obtained with 34 University of California — Experiment Station each of the types of longer pruning every year. The regularity of production in this case has brought about maximum yields of fruit of higher quality. TABLE 15 The Effect of Normal Pruning and of Longer Pruning with Thinning on the Yield of Molinera Types of pruning and thinning Years Normal no thinning Long-spur, with thinning Half-long, with thinning Cane, with thinning Crop to a vine 1925 pounds 10.7 12.2 14 1 46.3 18.0 20.3 pounds 24.2 26.6 29.2 26.6 pounds 20.3 25.9 30.8 30.6 32 • 27.9 pounds 23.8 1926 26.8 1927 31.2 1928 37.1 1929 34.2 306 The improvements in the quality of the fruit as a result of the longer pruning with thinning to regulate crop as compared with normal pruning are shown in table 16. TABLE 16 The Effect of Longer Pruning on the Fruiting of the Molinera; Averages for Four Years Type of thinning Clusters Normal berries Weight of 100 berries Type of pruning Weight Length Number to a cluster Per cent of total Germina- bility of pollen None pounds 1.01 1.10 1.36 124 0.98 inches 8.0 11.0 11.0 11.2 8.3 95 104 131 120 80 per cent 82 88 87 92 85 pounds 0.85 0.92 0.89 0.91 0.86 per cent 18 Long spurs Flower cluster 25 Half-long Flower cluster 28 Cane 37 Cane 17 The figures of table 16 indicate a considerable improvement in all of the factors measured, except weight of berry, over normal prun- ing, as a result of longer pruning when the crop was regulated by flower-cluster thinning. There was little difference in the improvement resulting from the different degrees of decreased pruning, except that the weight of cluster and number of normal berries to a cluster with cane and half-long pruning was somewhat greater than with long- Bul. 519] Pruning and Thinning Experiments with Grapes 35 spur pruning. The increase in weight is the direct result of the setting of the larger number of normal berries per cluster with the longer types of pruning. The greater increase in weight of cluster with the longer types of pruning is of questionable benefit, since the clusters of this variety are rather large with normal pruning. The fruit of the long-spur pruning with flower-cluster thinning to regulate crop was undoubtedly of the highest quality. The weight of cluster and number of normal berries to a cluster were increased only a little over normal pruning ; however, the length of cluster, the percentage of normal berries, and the weight of berry was increased almost as much as with cane pruning. These clusters were large, well filled with large normal berries, and with only a small percentage of shot berries. The fruit of the cane-pruned vines which were cluster-thinned after the fruit had set was no better than that of the normally pruned vines. Owing to their restricted number the berries were large, but the clusters were too straggly. Decreased pruning with thinning to regulate crop has increased the capacity of the Molinera vines for growth and production. Although the vines with longer pruning have borne larger and more regular crops as well as crops of better fruit, they have made more growth and their fruit has contained more sugar. This beneficial effect of longer pruning with thinning on the capacity of the vines is shown in table 17. TABLE 17 The Effect of Longer Pruning on the Increase in Area of Trunk, Weight of Prunings, and Sugar Content of the Fruit Juice of Molinera; Averages for Four Years Type of pruning Annual increase in cross-section- al area of trunk Annual weight of prunings Sugar content of the fruit juice Normal squart inches 0.73 .76 .78 0.81 pounds ' 7.9 8.2 8.1 8 5 ° Balling 20 22 21 Cane 21 As indicated by the figures of table 17, each of the types of longer pruning has resulted in increased trunk growth and a slightly greater weight of prunings. These differences, although small, are of impor- tance insofar as they help to show that larger crops and crops of better fruit can be produced on vines with longer pruning without injury to the vines if the crop is properly regulated. 36 University of California — Experiment Station THE RESPONSES OF VARIETIES THAT USUALLY SET WELL- FILLED CLUSTERS OF NORMAL BERRIES Although it is the rule for the clusters of the varieties discussed below to be well filled, the density or compactness of the clusters of the several varieties varies considerably. Some of the varieties are characterized by loose clusters, others by well-filled clusters, while still others regularly produce clusters which tend to be compact. In order to give as accurate an indication as possible of the response to longer pruning with thinning of the several varieties studied, each variety will be discussed separately. This should also give a better indication of the probable utility of these methods with other varieties which give similar fruiting responses with normal pruning. Malaga. — Beginning with the 1925 season 20 vines of Malaga have been cane pruned, 10 half -long pruned, and 20 normally pruned. In 1927 the number of cane and half-long pruned vines was increased to 35 and the number of normally pruned vines to 55. The crop on the cane-pruned and half-long pruned vines was regulated by flower- cluster thinning. During the season of 1928 the tests were extended to include two lots of 17 vines each of Malaga which were pruned to long spurs, one lot which was pruned half-long, and two lots which were normally pruned. The crop of the half -long pruned lot and one lot of each of the others was regulated by cluster thinning after bloom- ing. That of the other lot pruned to long spurs was regulated by berry thinning. The results obtained with the three types of longer pruning and one lot of normal pruning with three types of thinning as compared to normal pruning without thinning are shown in table 18. The figures of table 18 show that the quality of the fruit of the Malaga at Davis was not improved by longer pruning with the regu- lation of crop by flower-cluster thinning. The considerable increase in the number of berries that set per cluster with these types of longer pruning and thinning over that of the normally pruned vines has caused a decrease in the size of berry as well as a poorer development of color. The increase of 15 to 20 per cent in weight of cluster and the increase of approximately 20 per cent in the number of berries to a cluster has made the clusters unwieldly in packing owing to both size and compactness. There was no increase in the percentage of normal berries to a cluster over that of the normally pruned vines. The increase in length of cluster of 13 and 15 per cent, respec- tively, for the half-long and cane pruning with flower-cluster thinning Bul. 519] Pruning and Thinning Experiments with Grapes 37 over that of normal pruning is of questionable importance. Where the clusters of Malaga tend to be small or where it is desirable to make a cluster pack this increase in length of cluster may be of con- siderable value, provided, of course, that it may be obtained without a material loss in size of berry or color development. Under most conditions where the Malaga produces very large clusters with normal pruning this increase in length would not be an improvement, espe- cially where it is accompanied by a considerable loss in size of berry. TABLE 18 The Effect of Different Types of Longer Pruning with Flower-Cluster, Cluster, and Berry Thinning on the Fruiting of Malaga; Averages for Four Years Type of thinning Crop to a vine, pounds Weight of cluster, pounds Length of cluster, inches Normal berries Weight Type of pruning Number to a cluster Per cent of total of 100 berries, pounds Vines nine years of age Normal (spur).... 43 43 51 1.04 1 19 1.25 9.0 10 2 10.3 134 162 163 88 88 88 0.77 73 71 Vines six years of age None 23 18 27 24 23 1.00 1.09 1.03 0.82 1.07 9.1 9.3 92 9.4 132 133 124 91 130 89 89 90 90 89 76 Normal (spur).... 82 83 89 85 Longer pruning with cluster thinning after the setting of the berries has overcome the objectionable increases in size of cluster and number of berries that followed flower-cluster thinning. By delaying the thinning until after blooming and setting, the number of berries to a cluster remains aproximately the same as with normal pruning. The slight differences in the number of berries to a cluster shown in table 18 are due directly to the size of the clusters removed at thin- ning. The same is true of the influence of cluster thinning on length of cluster. The fruit on the cluster-thinned vines, however, was of superior quality to that of the unthinned vines in several respects. It was more uniform in size of cluster and size of berry. The berries with the several types of priming on which crop was regulated by cluster thinning were 8 per cent larger on the normally pruned, 9 per cent larger on the long-spur pruned, and 12 per cent larger on the 38 University of California — Experiment Station half-long primed vines, respectively, than those of the normally pruned vines without thinning. The color of the fruit of the vines on which the crop was regulated by cluster thinning also was more uniform and attractive. By removing the excessively large or compact and the small and misshapen clusters by cluster thinning, the energy of the vine was directed into clusters all of which were potentially marketable. Thus, a considerably higher percentage of the fruit of the cluster-thinned vines was suitable for packing than that of the vines normally pruned without thinning. Despite the considerable reduction in total crop on the normally pruned vines as a result of the cluster thinning, the amount of marketable fruit produced by these vines was almost equal to that of the vines similarly pruned without thinning. The vines pruned to long spurs or pruned half -long and cluster thinned not only produced a greater total weight of fruit than the normally pruned vines without thinning but also the fruit was of higher quality. By using berry thinning to regulate crop on the vines pruned to long spurs the quality of the Malaga fruit was somewhat better than that of similarly pruned vines on which the crop was regulated by cluster thinning. With berry thinning the average increase in weight of berry over that of the normally pruned vines was 17 per cent. The color development was better too. At Davis, however, where the Malaga is very vigorous and where the fruit usually remains grass green, the improvement in color was not sufficient to make the fruit of this variety of the best quality. Under more favorable conditions for the production of the Malaga, berry thinning may result in a greater improvement in the color of the fruit. This has been indicated in a semicommercial experiment made near Cutler in Tulare County, where the influence of berry thinning on coloring was sufficient to be of considerable practical importance. Owing to the increase in capacity for production as a result of longer pruning with thinning to regulate crop, the longer-pruned Malaga vines made as much or more growth and produced fruit of as high sugar content as the normally pruned vines, even though their annual average crops were larger. This beneficial effect of longer pruning on the capacity of the vines is shown in table 19. The figures of table 19 show that the vines of longer pruning made at least as much vine growth as the normally pruned vines. Although the differences are relatively small they are, nevertheless, of considerable importance in that they further substantiate the possibility of producing somewhat larger as well as better crops of fruit on these vines. Bul. 519] Pruning and Thinning Experiments with Grapes 39 TABLE 19 The Effect of Longer Pruning with Thinning on the Increase in Area of Trunk, Weight of Prunings, and Sugar Content of the Fruit Juice of Malaga ; Averages for Four Years Type of pruning Annual increase in cross section- al area of trunk Weight of annual pruning Sugar content of the fruit juice Normal square inches 0.47 0.47 0.48 0.49 pounds 4.76 4.77 4.95 5.85 ° Balling 22 23 Half-long 24 22 Emperor. — The Emperor pruning- tests were started on four-year- old vines in 1925. Half-long, cane, and normal pruning were em- ployed. In view of the fact that the basal buds of the Emperor are not usually fruitful, normal pruning in this variety means the reten- tion of spurs that are three to four and occasionally five buds long, the length of the spur varying in accordance with its capacity. 8 In these tests only 10 vines were used with each of the above types of pruning. The crop of the half -long and cane-pruned vines was limited by flower-cluster thinning. In 1927 another series of two lots of 35 three-year-old vines each were brought into the pruning tests. One of these lots was cane pruned and the crop limited by cluster thinning after blooming, and the other was normally pruned. TABLE 20 The Effect on the Fruiting of Emperor of Different Types of Longer Pruning with Flower- Cluster and Cluster Thinning; Averages for Three Years Type of pruning Normal (long spurs) Half-long Cane Normal (long spurs) Cane Crop to a vine Weight of cluster Length of cluster Normal berries Type of thinning Number to a cluster Per cent of total None pounds 25 26 32 15 21 pounds 1.03 1.29 1.34 1.00 1.03 inches 10.6 12.3 12.3 10 2 10.5 114 149 148 120 108 per cent 89 90 91 88 89 Flower cluster Flower cluster None Cluster Average weight of 100 berries pounds 0.89 85 0.84 0.84 0.96 s Athough the Emperor is now generally pruned to several long spurs with usually two short canes, the above method of pruning was in general use and hence recognized as the method of normal pruning when these tests were started. This change, which came about independently of the work reported here, repre- sents an application of longer pruning for the purpose of overcoming light bearing. It has materially improved the regularity of cropping and the quality of the fruit. 40 University of California — Experiment Station The response of the Emperor to the various types of pruning and thinning are shown in table 20. Table 20 shows that with half -long and cane pruning and flower- cluster thinning the clusters of Emperor were heavier, longer, and contained more berries than those of the normally pruned vines. Where the Emperor tends to produce small loose clusters with normal pruning or where it is desired to produce Emperors for cluster pack- ing, this increase in size of cluster may be of considerable value. At Davis, however, the Emperor tends to produce large, well-filled clusters with normal pruning, hence the increase in size following longer pruning with flower-cluster thinning has made the cluster very large and compact. Here the considerable increase in the number of berries to a cluster has resulted in a reduction in weight of berry of 5 and 6 per cent, respectively, for the half -long and cane pruning. The per- centage of normal berries to a cluster was influenced only very slightly by the flower-cluster thinning. Under the conditions at Davis cluster thinning to regulate crop has shown more promise than flower-cluster thinning in the improve- ment of the quality of Emperor. With cluster thinning, the size of cluster and number of berries to a cluster was changed very little, since the thinning was not done until after the berries had set. Here the Emperor tends to produce rather large clusters, hence in cluster thinning many large clusters were removed along with the misshapen and very small ones. This accounts for the smaller number of berries to a cluster on the cluster-thinned as compared with those on the normally pruned vines. Despite this decrease in number of berries to a cluster, the 14 per cent increase in weight of berry was sufficient to keep the weight of cluster of the thinned vines as great as that of the clusters of the normally pruned vines. The greater uniformity in size of cluster and size of berry, together with a more uniform development of color, has added much to the quality and appearance of the fruit. In addition to producing better fruit the vines with longer pruning and cluster thinning produced more fruit than the normally pruned vines. This increase in the amount of fruit produced by the vines with longer pruning and cluster thinning has resulted largely from the production of good crops each season rather than the production of any unusually large crops. The longer pruning has insured the setting of an excess of clusters for a full crop each year. Then by thinning after blooming it was possible to retain a sufficient number of good clusters to give the vines all the crop they could properly mature every year. With normal pruning, on the contrary, although these Bul. 519] Pruning and Thinning Experiments with Grapes 41 vines were pruned to long spurs, the Emperor has set very light crops in some seasons. Furthermore, the early removal of the excessively large and the small and misshapen clusters prevented the expenditure of energy by the vine on clusters that were incapable of developing into marketable fruit. This procedure alone very much increased the percentage of marketable fruit produced to a vine. Along with the larger crops of better fruit the longer-pruned vines have made a slightly greater vine growth each season than the nor- mally pruned vines. In the case of the half-long pruning with flower- cluster thinning to regulate crop, the increase in vine growth over normal pruning amounted to approximately 10 per cent. The sugar content of the fruit of the longer-pruned vines with cluster thinning was equal to or slightly higher than that of the normally pruned vines. Ohanez (Almeria). — The pruning tests with the Ohanez were started in 1925 on four-year-old vines. Ten vines each were employed with cane, half-long, and normal pruning. The crop on the longer pruned vines was regulated by flower-cluster thinning. The results are shown in table 21. TABLE 21 The Influence of Cane and Half-long Pruning with Flower-Cluster Thinning on the Fruiting of Ohanez ; Averages for Three Years Type of thinning Crop to a vine Weight of cluster Length of cluster Normal berries Type of pruning Number to a cluster Per cent of the total Weight of 100 berries Normal (spur) None Flower fluster Flower cluster pounds 24 28 31 pounds 74 1.00 1.08 inches 6.6 9 9.5 112 131 130 per cent 82 89 92 pounds 0.66 Half-long 0.77 83 Each of the measurements in this table shows an improvement in fruiting as a result of the longer pruning. This is of especial interest when it is considered that unsatisfactory fruiting has been the rule with this variety in nearly all of the locations in which an attempt has been made to grow it in California. The most important increases obtained by longer pruning with flower-cluster thinning of the Ohanez are those in number of normal berries to a cluster and length of cluster. With normal pruning the clusters of this variety tend to be small and short or globular, a shape and size which make them unsuitable for packing in sawdust. The increase in length seems of particular importance, since it adds more to the quality of these grapes for sawdust packing. The increase in length of cluster over that of normal pruning was 37 per cent for 42 University of California — Experiment Station half -long pruning and 44 per cent for cane pruning. This increase in length is a direct result of an increase in the ratio of leaves to flower clusters before blooming. The greater increase in the case of the cane- pruned vines was due to the higher ratio cf leaves to flower clusters with this type of pruning. The increase over normal pruning in the number of normal berries to a cluster was 17 per cent for half-long pruning and 16 per cent for cane pruning. The increase in the number of berries to a cluster on the longer- pruned vines, together with the increase in berry weight (17 per cent for half-long pruning and 25 per cent for cane pruning), have made these clusters much heavier. The clusters of the half-long-pruned vines were 35 per cent and those of the cane-pruned vines 46 per cent heavier than the clusters of the normally pruned vines. Accompanying the increases in the weight of cluster and berry, the length of cluster, and the number and percentage of normal berries to a cluster, there w^as an increase in the crop to a vine with longer pruning. With the half -long pruning this larger yield amounted to 17 per cent and with the cane pruning to 30 per cent. That the above beneficial influence on quality as a result of longer pruning with thinning has in no way injured the vines is shown by the data on growth in table 22. TABLE 22 The Effect of Decreased Pruning on the Increase in Area of Trunk. Weight of Prunings, and the Sugar Content of the Fruit Juice; Averages for Four Years Type of pruning Annual increase in cross section- al area of trunk Weight of annual prunings Sugar content of the fruit juice Normal square inches 0.48 0.48 58 pounds 5.1 5 5 6 6 Balling 22 Half-long 24 Cane 24 As indicated by the figures of table 22, the growth of the long- pruned Ohanez vines was equal to or superior to that of the normally pruned vines. In fact, the cane-pruned vines made somewhat over 20 per cent more growth than the normally pruned vines. There can be little doubt, therefore, concerning the beneficial influence of longer pruning on capacity, especially in view of the fact that these cane- pruned vines were producing crops 30 per cent greater in weight than the normally pruned vines. As shown by table 22, the fruit of the longer-pruned vines was higher in sugar content. Not only was this fruit sweeter at the time Bui,. 519] Pruning and Thinning Experiments with Grapes 43 of harvesting but it began to ripen earlier. The earlier commencement of ripening appears to favor color development, hence at harvesting a considerable percentage of the fruit of the longer-pruned vines showed a pink tinge or blush, while practically none of the fruit of the nor- mally pruned vines showed any. Tokay. — Of the varieties worked with, the Tokay has the greatest tendency to set dense or compact clusters. Except in a few limited areas this variety sets too many berries to a cluster wherever it is grown in California. In view of this condition, flower-cluster thinning, which improves set, was not employed as a means of regulating crop with the types of longer pruning on this variety. Moreover, as the Tokay also is a very regular bearer it was not necessary to greatly increase the amount of wood retained with the longer types of pruning to insure the setting of sufficient clusters for a full crop after the undesirable fruit was removed at thinning. Although the retention of more wood would increase capacity, it would tend to unduly increase the cost of thinning. Thus, only long-spur and normal prun- ings were employed. Since 1927 three lots of 17 vines each have been pruned to long spurs and one lot of 17 vines has been normally pruned. The three lots of vines pruned to long spurs were berry thinned to regulate crop. One lot was thinned as soon as the drop of berry forms following bloom was completed. The second lot was thinned 8 to 10 days later and the third lot 15 to 20 days later. The lot of normally pruned vines were not thinned. During 1928 and 1929 these tests were replicated three times in the Tokay section near Lodi. Since a report of the berry-thinning investigation has been pub- lished in a bulletin, 9 only a general account of its benefits will be included here. The methods of berry thinning and recommendations for their use were described and outlined in the bulletin referred to. In 1928 two other lots of Tokay vines pruned to long spurs were brought into the tests. One of the lots of vines was permitted to bring to maturity all of the clusters that set; that is, the crop was not regulated by thinning. The crop of the other lot of long-spur pruned vines was regulated by cluster thinning. The response of the fruiting of the Tokay to long-spur pruning without thinning, with cluster thinning, and with berry thinning as compared to normal pruning is shown in table 23. The averaged results for all of the berry-thinned plots for the three different dates of thinning are given. 9 Winkler, A. J. Berry thinning of grapes. California Agr. Exp. Sta. Bui. 492:1-22. 1930. 44 University of California — Experiment Station As indicated in table 23, long-spur pruning without thinning to regulate crop has resulted in the production of fruit of poorer quality than that of the normally pruned vines. This treatment has increased the amount of crop to a vine so much that the clusters were shorter and of less weight than those of the normally pruned vines even though the number of berries was approximately the same. The decrease in length of cluster parts made the clusters compact, although the berries were 21 per cent smaller than those of the normally pruned vines. As a result of the compactness of cluster and the relatively low sugar content, the coloring of this fruit also was very poor. TABLE 23 The Influence of Long-Spur Pruning with and Without Thinning on the Fruiting of Tokay Type of pruning Type of thinning Crop to a vine Average weight of cluster Number of berries to a cluster Average weight of 100 berries pounds 24 33 26 25 pounds 1.19 0.94 1.15 0.85 120 118 112 71 pounds 1.00 0.79 Cluster 1.03 1.20 Long-spur pruning with cluster thinning to regulate crop showed some improvement in the quality of the fruit. This was owing largely to the possibility of removing excessively large and compact as well as very small or misshapen clusters. The average number of berries to a cluster was reduced somewhat by the removal of oversize clusters. The thinning improved the size of berry slightly, so that the weight of cluster has been reduced very little. The fruit of the cluster-thinned vines showed a slightly higher sugar content at harvesting than that of the normally pruned. There was, however, no appreciable improve- ment in color. Thus, it appears that cluster thinning of the Tokay at Davis does not change the character of the clusters sufficiently to permit of any considerable improvement in quality. The clusters were still compact although the densest ones had been removed at thinning. In districts where the clusters of the Tokay are less prone to be compact, on the other hand, cluster thinning may improve the fruit. Long-spur pruning with berry thinning to regulate crop, on the contrary, has very greatly improved the quality of the Tokay, both at Davis and at Lodi. This improvement, no doubt arises largely from an alteration in the character of the clusters by the removal of cluster parts by berry thinning. The most important change is the reduction in the number of berries which makes the clusters less compact. The Bui,. 519] Pruning and Thinning Experiments with Grapes 45 removal of 40 per cent of the berries improved the nutrition of the retained berries very considerably. This improvement in the nutrition is indicated in part by the influence of the operation on size of berry. The average increase in size of the thinned berries over that of the berries of the unthinned normally pruned vines for the three dates of thinning for the several years was about 20 per cent. The earliest thinning resulted in an increase of berry weight of 31 per cent. When the thinning was delayed 8 to 10 days, the gain was one-third less; and when delayed 15 to 20 days, it was two-thirds less. This marked increase in size of berry has tended to compensate to a considerable extent for the decrease in number of berries as a result of thinning. Thus, a reduction in the number of berries of over 42 per cent has reduced the weight of cluster only 29 per cent. The fruit on the vines pruned to long spurs with berry thinning to regulate crop matured earlier than that of the normally pruned vines. In 1928 and 1929 the berry-thinned fruit was 3° Balling higher than that of the normally pruned vines at the first picking and 1.5° Balling higher at the second picking. Of possibly even greater importance than the increase in size of berry and earlier maturing following berry thinning, was the earlier and more uniform development of color by the thinned fruit. All of the fruit was picked for color, hence the amount of fruit removed at the several pickings gives an indication of the time of color develop- ment. In 1928 twice as much fruit was removed from the thinned as from the unthinned vines at the first and second pickings. In 1929 over one-half more fruit was removed from the thinned than from the unthinned vines at the first picking. The color not only developed earlier on the thinned fruit but it was more uniform; that is, there was less difference in the shade or intensity of color on the exposed and protected sides of the clusters. This more uniform coloring added greatly to the quality of the fruit and increased the rate at which the thinned fruit was removed from the vines at harvesting from 20 per cent faster to several times as fast. This increase in the rate of harvesting was due to the fact that the pickers could tell by a glance at one side of a thinned cluster whether it was ready for harvesting. With the unthinned fruit, especially where the clusters were large and compact, it was necessary to see both the exposed and the protected sides of a cluster before deciding. 46 University of California — Experiment Station DISCUSSION AND RECOMMENDATIONS By separating the influence of pruning from the influence of crop on the growth and fruiting of the vine, new principles of the response of the vine to pruning have been discovered. It has been definitely shown that pruning depresses the capacity of the vine for both growth and production. By employing longer types of pruning, i.e., by retaining a larger part of the yearly growth and accompanying this with some method of thinning to regulate the crop, a method of pro- cedure has been developed which offers promise for the improvement of both quality and yield in many varieties of table grapes. In adapting the benefits of longer pruning to practice it must be remembered that although the response of all varieties to pruning, other conditions being- equal, is the same, not all varieties fruit uniformly well under any one system of pruning. Thus an increase in capacity will be obtained with all varieties when more wood is retained at pruning so long as the crop is properly regulated by thinning, but the degree of longer pruning which will result in maxi- mum improvement in quality and yield consistent with economical practices will not be the same for all varieties. For example, in these tests the most satisfactory improvements in Muscat and Hunisa were obtained by the longest (cane) pruning practicable under present systems of vineyard management, while Malaga and Tokay did not respond economically to pruning which was longer than long-spur pruning. These differences in response to longer pruning by different varieties are connected with the set of their fruit with normal pruning. When the set of fruit approaches perfection with short pruning, as is the case with Malaga and Tokay, its quality cannot be improved appreciably by much longer pruning. With such varieties it is only possible to obtain the improvements in quality which result from greater uniformity or reduction in compactness. This can be accom- plished by a very moderate increase in the amount of wood retained at pruning, which increases slightly the amount of fruit set and thus allows for the removal of undesirable clusters or cluster parts by thinning. On the other hand, when the fruit of a variety is very poor in quality with normal pruning owing to an imperfect nutrition of the flowers, which causes the clusters to be straggly or results in the setting of many shot berries, as is the case with Muscat and Hunisa, relatively long pruning may prove to be economical. Such pruning Bul. 519] Pruning and Thinning Experiments with Grapes 47 brings about a large increase in the number of leaves early in the season which, when accompanied by the proper type of thinning (flower-cluster), increases the ratio of leaves to flower clusters so that the nutrition of the flowers is improved. Between the extremes in the perfection of the setting of fruit of the varieties just referred to there are other varieties which take an intermediate position with respect to the setting of their fruit with normal pruning. The factor of light bearing due to inherent characteristics such as small clusters and imperfection or rarity of blossoms also has sufficient importance in some varieties to influence the length of pruning. The response of the fruiting of the vine to thinning offers even greater possibilities in the improvement of quality than longer prun- ing. Longer pruning simply increases capacity by permitting the vine to produce its leaves earlier in the season and in greater numbers and requires completion by some form of crop restriction without which it is dangerous both to the quality of the fruit and the health of the vine. Suitable thinning is not only a necessary compliment of longer pruning but is of value in many cases where longer pruning is impossible or unnecessary. By thinning the character of the fruit can be very materially improved. The regulation of crop by thinning merely consists in limiting the crop sufficiently by removals so that the vine is capable of bringing it to perfect maturity. The choice of the method of thinning to be employed in regulating crop depends entirely upon the type of fruit a variety produces ; that is, on whether or not it is desirable to retain the character of the clusters as they normally develop or to change their character with respect to size, shape, or density. The utility or advantages of the different methods of thinning for the purposes just referred to can be best shown by a discussion of the methods of thinning and by indicating the time at which each is carried out and how it influences the character of the fruit. Flower-Cluster Thinning. — The clusters of rudimentary flowers of the grape come out with the leaves when growth starts in spring. The parts of the individual flowers, such as the pistil, anthers, petals, and sepals, are formed after leafing out and develop up to the point of blooming in from 6 to 8 weeks, according to the season. Now, if some of the flower clusters are removed early in this period, between leafing out and blooming, without disturbing the number of leaves on the vine, the retained clusters have a larger number of leaves to nourish them. With the increase in the ratio of leaves to flower clusters the flowers are better supplied with the food materials — carbohydrates — 48 University of California — Experiment Station that are manufactured in the leaves. As a result of the more abundant supply of carbohydrates, the flower parts develop more perfectly with a consequent better set, as well as an increase in the percentage of normal berries. The usefulness of flower-cluster thinning in the improvement of quality, therefore, is limited to varieties that have loose or straggly clusters or set shot berries, or both, with normal pruning. It will be most successful in overcoming these difficulties when it is carried out as soon as possible after all the flower clusters have appeared so as to obtain the maximum influence of the improved nutrition on the retained clusters. The proper stage of development Fig. 10. — A single cane of a Muscat vine showing the proper stage of development for flower-cluster thinning. for flower-cluster thinning is shown in figure 10. The earlier the thinning, as long as all of the clusters are out, the less expensive it is, since fewer of the clusters are obscured by the leaves. Cluster Thinning. — Cluster thinning means the removal of clusters after blooming, that is, after the berries have set. As it is not carried out until after blooming, it does not influence the number of berries to a cluster, and since entire clusters are removed it does not change the character of the retained clusters. Its usefulness in the improve- ment of quality, therefore, is limited to varieties that set well-filled clusters. This method of thinning results in an improvement in quality primarily through its influence on uniformity in size of cluster, size of berry, and coloring. The greater uniformity results from the removal of oversize, undersize, and misshapen clusters. Bxtl. 519] Pruning and Thinning Experiments with Grapes 49 Cluster thinning should usually be carried out as soon as all of the berries that will set have set. The stage of growth at which cluster thinning should be done is shown in figure 11. The early removal of undesirable clusters improves the conditions of nutrition for the retained clusters and hence the fruit will develop better. This is especially important as regards the increase in size of berry. Early thinning also prevents the wasting of energy by the vine in the devel- opment of clusters which are to be removed. Fig. 11. -A long-pruned vine showing- the proper stage of development for cluster thinning. Berry Thinning. — As used in California berry thinning consists in the removal of parts of the cluster. It usually consists of cutting off the end of the main stem (rachis) and several branches of the cluster or cutting off a sufficient part of the main stem so as to retain only the desired number of berries. Berry thinning can improve quality only when the berries are so numerous as to make the clusters too compact to be of high quality or when the cluster parts are so large as to interfere with proper coloring and maturing. Its usefulness in the improvement of quality, therefore, is limited to varieties that set very compact or very large clusters. Berry thinning usually changes the character of the clusters very materially. It always reduces their size and may alter their shape. 50 University of California — Experiment Station Berry thinning" should, be done as soon as possible after all the berry forms that normally drop off after blooming have dropped. Thinning at this stage of development results in maximum increases in size of berry. When the thinning is delayed 8 to 10 days the gain Fig. 12. — A method of berry thinning adapted to clusters with large shoulders, such as the cluster A. B, the shears placed for cutting the main stem of the cluster; C, the thinned cluster. (The berry thinning should be completed before the fruit reaches this size.) (From Bui. 492.) Fig. 13. Berry thinning by which the branches along one side, except the uppermost branch of the main stem, are removed and the end of the cluster shortened back. This method is adapted to long narrow clusters, such as that shown at A. B, the act of thinning, using a knife; C, the thinning cluster. (From Bui. 492.) But,. 519] Pruning and Thinning Experiments with Grapes 51 in berry size will usually be about one-third less ; and when delayed 15 to 20 days it will be two-thirds less. The proper stages of develop- ment for berry thinning together with different methods of thinning are shown in figures 12, 13, and 14. Fig. 14. — Another method of berry thinning adapted to long narrow clusters. This method retains the normal shape of the cluster. The positions at which cluster parts would be removed are indicated by the curved marks. (This stage of development is about one week too early for berry thinning.) (From Bui. 492.) Recommendation for Longer Pruning with Thinning. — Adapta- tions of longer pruning and thinning for the eight varieties used in these tests are summarized in table 24. This table is based on the results obtained at Davis with one or more of the types of longer pruning and when one or more of the three methods of thinning to regulate crop were used on each variety. Although half -long pruning with the several methods of thinning showed improvements which in some cases were somewhat greater 52 University of California — Experiment Station than those obtained with long-spur pruning and in others results which were only a little less than those obtained with cane pruning, it will not be considered in the recommendations owing to the difficulties of maintaining the desired form of the vines when this type of prun- ing is used. TABLE 24 Suggestions for Longer Pruning with Thinning to Eegulate Crop Variety Type of pruning Method of thinning Cane Cane Cane Long spur Long spur , Cane Cane Long spur Flower cluster Hunisa Flower cluster Dattier (de Beirut) Flower cluster Cluster (or berry) Cluster (or flower cluster) P'lower cluster, with berry in some seasons Tokay Berry (or cluster) The varieties Muscat and Dattier are subject to both couture and the setting of shot berries to a very marked degree. Hunisa is equally subject to the setting of shot berries. In these varieties, therefore, the largest increase in leaf area possible in the early part of the season, together with an early reduction in the flower clusters so as to increase the ratio of leaves to flowers, has resulted in the most satisfactory improvements in quality. This being the case, cane pruning with flower-cluster thinning appears to give most promise as a method of procedure to improve the fruiting of these varieties. Results obtained in Tulare and Madera counties (see table 10) for the Muscat, and in Tulare County for the Hunisa confirm this recommendation for these two varieties. Long-spur pruning with flower-cluster thinning has improved the fruiting of the Muscat a little, yet it is not believed that the small increase with this type of pruning is as economical as the larger increase obtained with cane pruning. The Dattier is somewhat less subject to coulure and the setting of shot berries than Muscat, hence in favorable locations long-spur pruning with flower-cluster thinning may result in a satisfactory improvement of its fruit. The Molinera (Red Malaga) has only a moderate tendency to set shot berries. It tends, however, to be a light and irregular bearer. Long-spur pruning has resulted in a sufficient increase in leaf pro- duction early in the season so that the tendency to set shot berries has been largely overcome by flower-cluster thinning. This type of pruning also retains a sufficient number of buds to insure the develop- ment of enough clusters to produce a full crop of fruit each year. Longer types of pruning and flower-cluster thinning with this variety have not resulted in a further reduction of the percentage of shot Bul. 519] Pruning and Thinning Experiments with Grapes 53 berries, but have tended to make the clusters too large and increased the cost of thinning. The Malaga at Davis usually sets clusters which have a large number as well as a high percentage of normal berries. Its fruit, therefore, cannot be further improved by operations, such as flower- cluster thinning, which tend to improve set, The quality can, however, be improved by means which promote greater uniformity in size of cluster, size of berry, and coloring. Long-spur pruning of Malaga has increased the number of clusters that develop sufficiently to permit weeding out the oversize, undersize, and misshapen clusters by cluster thinning. When the clusters of Malaga tend to be very large the quality of the fruit can be improved somewhat more by berry thinning than by cluster thinning. The berry thinning under such conditions results in a greater increase in the size of berry and in a better coloring of the fruit. Flower-cluster thinning to regulate the crop of Malaga should only be attempted in those localities where the clusters are normally loose or where it is desired to produce very large clusters. As the Emperor tends to be a light bearer, it is ordinarily pruned to long spurs with usually two short canes. Its clusters with this type of pruning usually set a high percentage of normal berries, In some localities the clusters tend to be rather loose. This latter condition is not the rule with Emperor at Davis. An improvement in the quality of the Emperor at Davis, therefore, is largely limited to an improve- ment in uniformity of size of cluster, size of berry, and coloring. Cane pruning with cluster thinning to regulate crop has shown the greatest promise with this variety. With cane pruning a sufficient number of clusters develop to allow the undesirable clusters to be removed at thinning. In localities where the Emperor clusters tend to be loose or when it is desirable to produce large clusters the crop may be regulated by flower-cluster thinning. This method of thinning im- proves set and since it is done before the leaves obscure the clusters it is the cheapest means, other than pruning, of regulating Grop. Unsatisfactory fruiting has been the rule with Ohanez in all of the localities of California where it has been grown. Its principal defects have been light bearing and a tendency to produce small clusters. Cane pruning has overcome the tendency to produce light crops and with flower-cluster thinning has sufficiently improved the nutrition of the flowers before blooming to greatly increase the length of the cluster. Set also has been improved. This type of pruning with flower-cluster thinning to regulate crop appears to offer real promise in improving the fruiting of this variety. In some seasons the clusters have set too many berries as a result of the flower-cluster thinning. 54 University of California — Experiment Station Since grape clusters do not increase in size after blooming, however, it is necessary to use flower-cluster thinning- to obtain clusters of large size. Where too many berries set on some clusters these may be reduced in number by berry thinning. Although the berry thinning adds to the cost, it further increases the size of berry, improves color- ing, and makes for a more attractive appearance, since it affords an opportunity for the removal of berries showing Ohanez spot. The Tokay usually tends to set too many berries to a cluster with all of the types of pruning used. To improve its quality it is necessary to alter the character of the clusters by the removal of cluster parts. This can be accomplished by long-spur pruning with berry thinning to regulate crop. The long-spur pruning allows for the development of a sufficient number of clusters to insure a full crop after thinning. By berry thinning enough of the berries (cluster parts) are removed to overcome the tendency of the clusters to be too compact. Where the Tokay clusters are relatively loose, some improvement in uni- formity, size of berry, and coloring can be obtained by long-spur pruning with cluster thinning to regulate crop. Other Varieties. — The type of longer pruning and method of thinning to regulate crop which are most likely to improve the quality of the fruit of any variety is best indicated by the character of the clusters of the variety with normal pruning. By comparing the fruit of the variety in question with that of the varieties employed in these tests it should be possible to devise a method of procedure which will be suitable. For example, a variety having straggly clusters or setting shot berries should be improved by the type of pruning and thinning suggested for Muscat ; a variety setting well-filled clusters of normal berries should respond like the Malaga; and a variety setting dense clusters of normal berries should respond like the Tokay. Two years' results with Cornichon and Olivette Blanche, varieties which are somewhat similar to the Emperor in their fruiting at Davis, indicate that they are most improved by the type of pruning and thinning suggested for the Emperor. The Gros Colman, which sets well-filled clusters of normal berries, has responded to longer pruning and thin- ning very much like the Malaga, In the case of very large-clustered varieties, such as the Thompson Seedless (Sultanina) the different parts of the clusters sometimes ripen unevenly. It often happens that the berries toward the base of the cluster are ripe enough to be eaten with satisfaction, while the berries at the end of the clusters are still sour. This condition is espe- cially objectionable in the early sections where it is to the interest of the grower to move his fruit as early as possible. This trouble may be largely overcome by either cluster or berry thinning. In either case Biru 519] Pruning and Thinning Experiments with Grapes 55 the vines are pruned somewhat longer to allow for the development of a sufficient number of clusters to have a full crop after thinning. Cluster thinning overcomes the uneven ripening of the excessively large clusters by removing such clusters entirely at the time of thinning. Where the entire removal of the very large clusters is impracticable their uneven ripening may be overcome to a large extent by removing the apical ends of the clusters or of long cluster parts. As a rule only the main stem of the cluster will need to be shortened, yet in the case of very heavy clusters it is worth while to remove also the ends of heavy and unusually long shoulders. In this thinning, one-fourth to one-third of the length of the cluster or cluster part is cut off, according to the size of the cluster or cluster parts and the number of clusters to a vine. When this thinning is done soon after blooming, the scars where the cuts were made heal completely and the retained berries tend to fill in over the cut ends so that the clusters have the appearance of being normal. Size of berry is increased and coloring is earlier and more uniform. COST OF LONGER PRUNING AND THINNING Longer Pruning. — The cost of longer pruning will vary with the increase in the amount of wood retained at pruning. The range of variation will be between the cost of spur pruning and cane pruning. Long-spur pruning costs but little more than normal spur pruning. The increase in cost in this case arises primarily from the greater care which must be exercised in the choice of the wood to be retained. Cane pruning with most varieties will cost approximately the same as the present pruning of Thompson Seedless (Sultanina). For cane pruning it is also necessary to erect a trellis for the support of the canes. Thinning. — Flower-cluster thinning will vary in cost according to the number of clusters removed, and the stage of development at which it is done. It can be done most cheaply when the clusters are entirely unfolded and before the leaves have developed very far. For the several seasons' work at Davis, the cost has averaged 2% cents to a vine for Muscat pruned to from five to seven canes which averaged 3 feet in length. In two test plots in Kern County with cane pruning the cost was 3 cents to a vine. With long-spur pruning, where fewer clusters would be removed, the costs would be less. From $5 to $15 an acre is probably the range of average cost. The cost of cluster thinning is somewhat higher than flower-cluster thinning for the removal of a similar number of clusters. The vines have made considerable growth by the time blooming is over, hence the clusters are not so readily seen. In the case of cluster thinning, 56 University of California — Experiment Station however, only enough extra wood is retained at pruning- to insure a full crop and to allow for the removal of undesirable clusters, hence the number of clusters to be removed should never be very large. The average cost per acre will range from $2 to not more than $10. Berry thinning will cost from $3 to $30 an acre according to the age and shape of the vines, the number of clusters thinned, and the method of thinning employed. With the simpler methods of thin- ning (see fig. 12) it will vary from $3 to $15 an acre for such varieties as the Tokay and Malaga, With the method of thinning shown in figure 14 it may be as high as $30 an acre. CAUTIONS Longer pruning of any variety that has been producing full crops will result in overbearing and the production of fruit of poor quality unless thinning is used to regulate the yield. This situation is due to the fact that more buds are retained with longer pruning and that the buds on the canes of practically all varieties become more fruitful from the base upward over the length usually retained in cane pruning. This great increase in the number of flower clusters tends to check shoot growth early, so that the production of leaves by the longer-pruned unthinned vines may even be less than that of the normally pruned vines on which crop is restricted by the pruning. The regulation of crop by thinning is a new operation to most grape growers. Until some experience is gained in thinning, too much fruit will frequently be retained on the vines. This is especially true of flower-cluster thinning, which is clone before blooming and which (if done properly) results in a considerable increase in the weight of the retained clusters. Overcropping with longer pruning indicates insufficient thinning. If the clusters are still too compact after berry thinning it simply shows that not enough berries were removed or that the part of the cluster that tends to become compact was not thinned sufficiently. Where large crops of fruit of very high quality are being pro- duced, a change to longer pruning with thinning would be of ques- tionable value. The only gain to be expected under such a condition would be an increase in regularity and amount of crop. Where the fruit of a variety is not of the desired quality or the crops are small or irregular with the present systems of pruning, a change to longer pruning with some method of thinning to regulate crop is justifiable. A change should, however, be made only after the grower is thor- oughly familiar with the extra cost and work which is required by the type of longer pruning and the method of thinning which he will employ.