UNIVERSITY OF CALIFORNIA COLLEGE OF AGRICULTURE AGRICULTURAL EXPERIMENT STATION BERKELEY, CALIFORNIA CONTROL OF THE MEALY PLUM APHID LESLIE M. SMITH BULLETIN 606 APRIL, 1937 UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA CONTENTS PAGE Annual cycle 3 Injury 5 Control 8 Elimination of secondary hosts 8 Fall defoliation 8 Method of measuring control in field plots 10 Early spray tests, 1928-29 12 Multiple applications 14 Laboratory tests of toxicity of several materials to mealy-plum-aphid eggs 17 Fall applications compared and combined with spring applications, with and without nicotine 19 Early field tests with coal-tar distillates 20 Tests with ten proprietary coal-tar sprays 23 Tests to determine the minimum effective concentration of tar distillate. . . 27 Discussion and conclusions 29 Emulsification 29 Concentration 29 Injury to the trees by coal-tar distillate 30 Stimulation of trees by tar sprays 30 Effect of tar sprays on men 31 Deciding whether to spray or not to spray 31 Other advantages derived from tar sprays 32 Supplementary methods of control 33 Summary . . . . ; 33 Acknowledgments 34 CONTROL OF THE MEALY PLUM APHID 1 LESLIE M. SMITH 2 The mealy plum aphid, Hyalopterus pruni (Geoff.) is distributed throughout the world aud is a particularly serious pest of plums and prunes in the north temperate zone. In subtropical regions it does con- siderable injury to apricots and peaches. In California, the mealy plum aphid confines its attack to blue plums and prunes, that is, Prunus domestica Linn. The present study of the control of this pest began in 1928 and has been continued through 1935. Field control experiments were conducted in Placer County on table-fruit varieties, and in San Joaquin County on drying or prune varieties. Before control studies can be properly presented, the complicated annual cycle of this pest must be reviewed ; this is done briefly in the following section. ANNUAL CYCLE 3 The mealy plum aphids pass the winter in the egg stage on the plum trees. The eggs (fig. 1) are found near the bases of the buds, and most commonly between the bud and the twig. They hatch in the spring shortly after the buds start to swell, and are practically all hatched by the time 5 per cent of the flowers are in bloom. The aphids which emerge from the overwintered eggs are known as the stem mothers. They are wingless, green in color, and match the color of the young leaves. Unlike the generations that follow, they are devoid of the white waxy powder which is characteristic of this species and which has given rise to its common name. All of the aphids from the overwintered eggs are females and give birth parthenogenetically (that is, without fertilization) to other wingless, parthenogenetic females. Several similar generations (3 to 7) occur on the plum trees, but with the advent of warm summer weather in June and July, the young, when matured, have wings and can no longer feed on the plum tree. These winged females fly to reed grass or cattails, where they give birth parthenogenetically to wingless aphids, identical in appearance to those 1 Received for publication July 21, 1936. 2 Junior Entomologist in the Experiment Station. 3 For a more complete discussion of the life history of this pest see : Smith, L. M. Biology of the mealy plum aphid, Hyalopterus pruni (Geoffroy). Hilgardia 10(7) : 167-211. 1936. Smith, Leslie M. Growth, reproduction, feeding, and wing development of the mealy plum aphid in relation to climatic factors. Jour. Agr. Research 54(5) : 345-64. [3] 4 University of California — Experiment Station of the spring generations on the plum. Here several generations of wingless females mature during the late summer and fall. In the latter part of October and throughout November, winged aphids are produced on the secondary hosts. They are unable to feed on the secondary hosts, after reaching maturity, and must fly back to the plum trees. The first winged aphids that are produced are all par- thenogenetic females, which after arriving on the plum give birth to young that mature to small, wingless, sexual females. The winged aphids Fig. 1. — Eggs of the mealy plum aphid laid around the bases of French prune buds. produced somewhat later on the secondary hosts are males, which fly to the plums and mate with the wingless sexual females. After mating, these females lay the overwintering eggs. The movement of aphids from tree to tree and from orchard to orchard is of primary importance in any consideration of control. Contrary to the popular belief, the winged aphids developed on the plum are in- capable of feeding on or producing young on the plum and must migrate to the secondary hosts. This has been demonstrated by repeated cage tests and by field evidence. In certain orchards the aphids are found year after year on certain trees while other trees are always free from aphids. This fact has given rise to the popular misconception that the aphids complete their annual cycle on the plum, and doubt is thereby thrown on the necessity for sec- Bul. 606] Control of Mealy Plum Aphid ondary hosts. In cases of this kind, it can generally be shown that the infested trees are near a windbreak, such as a barn or row of tall trees. The aphids, returning from the secondary hosts in the fall, settle on trees in such sheltered situations. Conclusive proof has been obtained that the aphids cannot complete their annual cycle without passing a part of the year on a secondary host. INJURY The injury done to plum and prune orchards by this pest may occur in three distinct forms : first, curling and stunting of the foliage, together with a reduced growth in length of new wood and a general stunting of the tree ; second, soiling of the surface and bloom of table fruits by aphid excrement and the frequently associated black sooty molds; and third, splitting of the fruit. Curled leaves may or may not accompany plum-aphid infesta- tions. If the number of overwin- tering eggs is large and the infestation in the early spring is consequently heavy, leaf curling is generally the result; but in light infestations or late heavy infestations, the leaves are not curled, since this aphid does not possess the ability to curl ma- ture leaves. Even though the leaves may not be curled, they are usually dwarfed by plum-aphid attack, and the dwarfing becomes more severe as the aphid population increases. After the aphids mi- grate from the trees, a few ter- minal, normal-sized leaves are usually developed during the fall growth period, as indicated in figure 2. The general stunting effect of these aphids was demonstrated in a test at San Jose. Twelve young prune trees were weighed and planted on January 15. During the following spring and summer, six trees were kept heavily infested and six were kept free from aphids by dusting. On February 4 of the following year, the trees were dug, washed, and Fig. 2. — Foliage of Grand Duke plum curled and stunted by the mealy plum aphid. The shoot on the left is not in- fested, in the center and right, previously infested. The normal terminal leaves were developed after the aphids left the tree. 6 University of California — Experiment Station weighed. The infested trees showed an increase in weight of 18.8 per cent, while the noninfested trees increased 25.9 per cent. This test indi- cated that the aphids caused a loss of 7.1 per cent of weight. Similar information was derived from control studies, reported later in this paper, in which old Diamond plum trees were sprayed in a checkerboard pattern with alternating nonsprayed blocks. Sixteen trees Fig. 3. — Increase in growth after an application of tar spray: A, unsprayed tree ; B, tar-sprayed tree. constituted a block. In general, the nonsprayed trees were heavily in- fested, while the sprayed trees were free from aphids. After the leaves had dropped, the sprayed trees showed more new wood than the non- sprayed trees, as shown in figure 3. Six sprayed blocks, with their associ- ated checks, were carefully pruned during the following winter. The lengths of the ten longest whips from each tree were recorded as well as the total weight of all of the prunings from each tree. These data were averaged for each block and computed into percentage increase over the adjacent check blocks. Considerable variation occurred in the results, Bul. 606] Control of Mealy Plum Aphid 7 largely because different sprays were used and some of the blocks were freer from aphids than others. The results are shown below : Per cent Per cent Sprayed block increase in increase in length weight 1 16.0 30 2 23.2 100 3 5.3 10 4 16.0 70 5 33.9 160 6 1.7 30 Average 15.9 66.7 The average increase of 15.9 per cent in length and 66.7 per cent in weight, however, is believed to be a conservative estimate of the retarda- tion which this aphid may produce. Fig. 4. — French prunes split as a result of mealy -plum-aphid infestation. The second type of injury produced by the plum aphid occurs when their excrement falls on the fruit and produces smears and spots. This is objectionable only in the case of fresh fruit; such smears are removed from prunes in the dipping process. All of the varieties of blue plums grown for the fresh-fruit market are covered with a white bloom. Aphid excrement destroys the bloom and produces unsightly blotches. Wash- ing or wiping the fruit cannot be practiced since the market demands fruit with the normal powdery bloom. Entire crops have been rendered unsalable by such soiling of the fruit. The third type of injury consists of cracks or splits produced in the flesh of the fruits. Such cracks are always on the lower end of the fruit 8 University of California — Experiment Station and are usually in the form of a straight line, a Y, or an X, as shown in figure 4. In the coastal valleys — Santa Clara and Sonoma — splitting almost always accompanies aphid attack, but it is entirely absent in the Sacramento and San Joaquin valleys. Conclusive evidence is at hand to prove that the splitting of the fruit is the result of aphid attack, since the fruit is frequently found to be split on an infested tree while an adjacent noninfested tree has no split fruit. More conclusive still is the fact that if only half of a tree is infested, the fruit on the infested por- tions generally splits, while that on the noninfested portion of the same tree does not. When very heavy infestations occur, these aphids cause the fruit to drop. In California, the infestations are generally not sufficiently heavy to produce this effect. During the last nine years this type of injury was observed only once : a French prune orchard with an abnormally heavy infestation dropped practically all of its fruit. CONTROL Elimination of Secondary Hosts. — Since the presence of secondary hosts is necessary to enable the aphid to complete its annual cycle, attempts were made to reduce the aphid population on the plum trees by the elimination of secondary host plants in the vicinity. At Davis, all of the secondary hosts within a radius of 4 miles of a plum orchard were cut below the water line and destroyed before any winged aphids had appeared, and hence before any return migration to the trees had oc- curred. During the season which followed this treatment, the plum orchard received one of the heaviest infestations ever observed in it; clearly the aphids migrated into the orchard in great numbers from secondary hosts greater than 4 miles distant. Subsequent observations in San Joaquin County indicated that the aphids were capable of migrat- ing 20 or 30 miles in numbers sufficient to heavily infest large acreages of plums. The removal of cattails and reed grass, therefore, is valuable as a means of control only when applied to wide areas. Fall Defoliation. — As indicated in the discussion of the annual cycle, the sexual females are borne on the plums in the late fall. They feed on the leaves of the plum for several days before reaching maturity and laying the overwintering eggs. If the trees were defoliated early enough, there would be no opportunity for the sexual females to feed. To test this theory, three trees were defoliated by hand-picking the leaves, on Octo- ber 16-17, before the return migration. In the following spring, these three trees were completely free from aphids. Hand-picking the leaves is a slow, tedious process. The time required Bul. 606] Control of Mealy Plum Aphid 9 to pick the leaves from a single unpruned Diamond plum tree was 3% hours. Since early defoliation was effective in controlling the aphid, and hand-picking was not economically feasible, attempts were made to de- foliate the trees with leaf -killing sprays. Commercial sodium nitrate was tested because this material had been shown, by other investigators, to kill leaf tissue when applied in concentrated solution ; and because it possesses value as a fertilizer. Tests were applied in 1931 and 1932, in San Joaquin and Placer counties. Five plots were sprayed, including 99 trees in all. Dilutions of % and 1 pound of sodium nitrate per gallon of water, were tested. Varieties of trees sprayed were French prune and Grand Duke and Diamond plums. The applications of sodium nitrate failed to control the aphids, chiefly because the defoliation was not complete. The applications on October 29 and November 1, 1931, were too late, since some eggs had been deposited at that time. Although the spraying was carefully done, there remained a number of small groups of leaves which were missed and hence green. The killed leaves remained attached to the trees until about the time of normal leaf fall, so that the aphids had opportunity to walk off and re- establish themselves on the green, unsprayed leaves. The spraying in 1932 was followed by a long period during which the humidity was low and little if any dew formed. Although the sprayed leaves were covered with crystals of sodium nitrate, no killing of the leaves occurred, and the sprayed leaves remained green until the time of normal leaf fall. At San Jose, during the fall of 1932, the effect of sodium nitrate on the trees was studied with particular reference to the possibility of push- ing the dormant buds when early defoliation was combined with irriga- tion. In every case the spray was commercial sodium nitrate at 1 pound per gallon, and the trees were sprayed heavily. The irrigation was by basin flooding. The trees were sprayed and irrigated as follows : Irrigated No Sept. 24 No Sept. 24 No No No No No On the first day after spraying, the leaves showed no injury, but by the third day many leaves were brown. The killing of the leaf appears Tree No. Sprayed 1 Sept. 10; Sept. 17; Oct. 2 2 Sept. 17; Sept. 24; Oct. 2 3 Sept. 24 ; Oct. 2 4 Oct. 2 ; Oct. 9 5 Oct. 9; Oct. 15 6 Oct. 15; Oct. 22 7 Oct. 22 8 9 Not sprayed Not sprayed 10 University of California — Experiment Station first along the leaf margin and progresses into the areas between the larger veins. Finally the entire leaf turns brown but remains pliable and leatherlike for some days. The fact that the brown leaves remain on the trees is inimical to the control of the aphid, since it affords opportunity for them to walk off and seek the few remaining green leaves. Thus, on tree No. 1, 20 days after spraying, only about 75 per cent of the leaves had dropped, and some green, evidently missed leaves were conspicuous. None of the trees, whether irrigated or not, pushed their dormant buds. The following spring, however, trees 1 and 2 showed a considerable num- ber of dead buds and twigs. Early fall defoliation by means of a leaf -killing spray of sodium ni- trate, therefore, does not seem advisable since : (1) a complete coverage, and hence killing of the leaves, is impossible, even with two applications ; (2) dry fall weather may render the spray ineffectual ; (3) if the sprays are applied sufficiently early to anticipate all egg deposition, injury to the tree may result. Method of Measuring Control in Field Plots. — In the 1928-29 tests, and in all others discussed subsequently, the same method of measuring the control was used. Each year all of the sprayed blocks, regardless of the time of year of application, were counted between May 10 and 20. The aphid infestation was assumed to be in the same stage of develop- ment in all of the blocks, between these dates. The blocks were counted as near simultaneously as possible. Counts of aphids were impossible since a single heavily infested leaf may support from 300 to 600 aphids. Consequently, trees were counted by classifying them into three degrees of aphid infestation. These types of infestation were (1) clean, in which case no aphids could be found on the tree; (2) lightly infested, in which case one, two, or three colonies were found on the tree ; (3) heavily infested, in which case four or more colonies were found on the tree. A colony generally consisted of the offspring of a fall migrant. In some cases more than one migrant may have contributed to the colony, which in part was responsible for a great variation in the size of colonies. A colony was generally confined to a single limb of the tree and was hence recognizable as a unit. Heavily infested trees usually were com- pletely covered with aphids so that no colonies, as such, could be recog- nized. From the practical or grower's standpoint, lightly infested trees may be considered as clean trees in the interior valleys, since the few aphids on such trees do little if any injury. The data in the following tables, comprised of counts made in this manner, should not be confused with data on the actual kill of aphids. Bul. 606] Control of Mealy Plum Aphid 11 73 d « fci 02 « Oh 02 ■J as -- o Si 55 ° T3 >> C3 Q o 9 ^ fcH o A o © o © o OS CO >— I O CO i-H CM »-H o> cm © CM 00 o 1-1 O -f o ■* CO o> © © CO m n us o 00 00 CM > o o > > o o > o > O o o a ft CM i-H > > IS (]> M M "o "o A A o o T3 >> 03 t- ft GQ O d ooo o d I M GO 0) O P3 w a d >> u a a> Ph d >> i- d a> Ph > Q E0 oj a M OJ O Tl O S />/1 &<§>>&» «, applied. These are : (1) the sexual /% - /», f\ #* ?** f\>> cycle or fall period, which occurs ^ generally in October and Novem- ^Ej ber; (2) the egg, which is on the 03 o> X co »0 o »fl if5 CO co 03 00 o O CO lO CO CO oo o >o o CO t- CO t--.-c iO cn« OS (M CN »0 o «o CO lO t^ tf5 tf5 O > at u a S3 artartrtartrtrt~ CM -Q • r~> CMCN fc, . co o o o o 73 co CD 0> — & ~ ' ■ CO CO 2 *"~ t~- ■* "IhS rt ,-h CM a -- to a a a a o «!«!«< «<;z; s s d as O bfl '-Do QSS rt M 03 rt rt rt a c3 a rt Ml W a. CD rt > o .d a rt o CO S3 jd rt d^ 01 o CD 03 rt §J CD o CD c3 a a o ED CD 03 CD Cj J3 CD >> rt t- a co O rt > •a c o s rt CD 3 Q -d s o -d a o S rt J3 o CI CD H J3 o d CD d d o O d a 03 d ° d ■g rt OQ CD d rt 16 University of California — Experiment Station have indicated that single applications are more effective when applied earlier. The value of earlier applications can be seen by contrasting the results obtained in the March 16 and 27 applications with those of March 27 and April 4, on both ranches 1 and 2. In these cases the timing of the sprays some 9 days earlier produced increases in clean trees of 20 per cent and 11 per cent, respectively. It has been shown (tables 2 and 6) that petroleum oils are only slightly toxic to the eggs of the mealy plum aphid. Spring sprays which are ap- plied before completion of the hatching period are, therefore, ineffective TABLE 5 Increase in Aphid Infestation in One Month, Placer County, 1929* Counted Per cent of trees infested Block Heavily Lightly Clean 1 May 11 31.8 76.1 20.6 85.3 20.6 69.8 24.3 77.1 60.9 17.4 52.9 11.8 53.8 18.6 55.9 15.9 7.3 June 11 6.6 2 May 11 26.5 June 11 2.9 3 May 11 25.6 June 11 11.6 Average May 11 19.8 June 11 7.0 * Data from ranch 4 of table 3. against those aphids which have not yet emerged from the egg. The young stem mothers, which hatch from the overwintered eggs, establish themselves on the green sepals of the unopened flower buds. Owing to this exposed position, and to the fact that their bodies are devoid of the water-shedding wax which protects all later generations, the stem mothers are most easily hit and wetted by the spray material. The most effective period for the application of spring sprays therefore occurs between the time of hatching and the opening of the flower buds. It has been found that practically all of the eggs have hatched by the time about 5 per cent of the flowers have bloomed in tlje case of most varieties of blue plums. This stage in the development of the tree may serve as an indication of the best time to spray. In order to be of value, any control of the mealy plum aphid must be very nearly complete. The reproducing females produce about ten young per day on the average. A control of 90 per cent of mature females would be more than compensated (theoretically) by the reproduction of the remaining 10 per cent during the first 24 hours after spraying. As has been stated earlier, counts were made on sprayed blocks from May 10 Bul. 606] Control of Mealy Plum Aphid 17 to 20. Subsequent multiplication of the aphids greatly increased the heavily infested trees, as shown in table 5, wherein two counts one month apart, on ranch 4 of table 3, are given. These data show that the heavily infested trees increased 217 per cent, and, as stated before, "heavily in- fested" trees are the only type which suffer economic injury in the interior valleys. Laboratory Tests of Toxicity of Several Materials to Mealy -Plum- Aphid Eggs. — During the winter of 1930-31, laboratory tests were con- ducted with various petroleum-oil sprays to determine whether or not TABLE 6 Relation of Viscosity and Unsulfonatable Residue to Toxicity to Aphid Eggs Number of eggs used 84. 86. 78. 107. 80. 72. 58. 75. 61. 55. 132. 92. 67. 98. 80. 91. 21. Sprays arranged in order of in- creasing viscosity Viscosity 50 55 60 70 70 75 83 102 110 110 125 130 140 Not sprayed No. 1 Not sprayed No. 2 Not sprayed No. 3 Not sprayed No. 4 Per cent of eggs dead 32.2 38.4 12.9 54.2 47.5 54.2 29.3 32.0 39.3 43.7 20.5 33.0 43.3 15.3 11.3 12.0 23.0 Sprays arranged in order of in- creasing unsulfonatable residue Unsulfonatable residue, per cent 53 70 75- 80 90 90- 92 90- 92 90- 92 90- 92 95-100 95-100 95-100 95-100 Not sprayed No. 1 Not sprayed No. 2 Not sprayed No. 3 Not sprayed No. 4 Per cent of eggs dead* 32.0 43.3 20.5 54.2 43.7 29.3 54.2 38.4 39.3 47.5 12.9 32.2 15.3 11.3 12.0 23.0 * Data in column 3 rearranged, with the 130-viscosity oil omitted, since the unsulfonatable residue of this oil was not known. viscosity or unsulfonatable residue were related to toxicity. Eggs were collected in the field on French prune whips, usually 2 to 3 feet in length. Sprays were applied February 13 to 16, with a small, low-pressure hand sprayer. Spraying was continued until the deposited droplets began to coalesce and run off. After spraying, all twigs were kept in a lath-house until after hatching was completed. Counts of hatched eggs were then made by dissecting each egg or shell with a needle, under a microscope. The data in table 6 were obtained. These counts fail to indicate a relation between toxicity and viscosity or unsulfonatable residue. If such a rela- tion exists, it occurs within a narrow range of efficiency and hence will require more refined technique for its demonstration. A comparison of the toxicities of paste emulsions and of miscible oils 18 University of California — Experiment Station was made in the same way. The results failed to show a consistently higher toxicity for miscible than for paste-type emulsions. The average percentage of eggs dead was, for miscible oils 42.1, for paste-type emul- sions 38.2. The low kill of eggs by petroleum-oil sprays that is indicated by laboratory tests is in conformity with results of field tests reported in table 2. A laboratory test on the relation of the percentage of oil in the mixed spray to toxicity was conducted with the same technique as before. Two proprietary winter petroleum-oil sprays were tested, each at 1, 4, and 10 per cent oil content. In all, 484 eggs were sprayed ; 290 unsprayed eggs TABLE 7 Toxicity of Materials Added to Oil Sprays Material Light oil Nicotine sulfate. Light oil . . . Pyrethrum. Light oil alone . Anthracene and oil mixture* . . , Dinitro-cresol and oil mixture* . Oil mixture Not sprayed Dilution, per cent 4.0 0.5 4.0 0.5 4.0 4.0 4.0 4.0 Number of eggs used 93 51 61 67 67 125 290 Per cent of eggs dead 77.4 47.1 39.3 40.3 92.5 24.8 13.7 * Prepared as described in text. constituted a check. The kill of eggs is summarized as follows : check, 13.7 per cent of eggs dead ; 1 per cent oil, 13.7 per cent dead ; 4 per cent oil, 42.0 per cent dead ; 10 per cent oil, 55 per cent dead. While these data indicate an increase in toxicity with an increase of oil content, the total kill produced by the 10 per cent mixture was still far from satisfactory, and again lends support to the findings in the field (table 2) . An attempt was made to increase the ovicidal action of petroleum-oil sprays by adding nicotine, pyrethrum, anthracene, and ortho-dinitro- cresol. Pyrethrum was obtained as a proprietary alcoholic extract. Dini- tro-cresol to the amount of 7.2 grams was dissolved in a mixture of 45 cc of light oil and 45 cc of kerosene, and emulsified. Anthracene was pre- pared as a saturated solution in a similar oil-kerosene mixture. Only a small amount of anthracene dissolved, however. The results are given in table 7. The toxicity of Penetrol and coal-tar distillate were likewise tested at 4 per cent. Of the eggs sprayed with Penetrol, 18.8 per cent failed to hatch ; while of those sprayed with coal-tar distillate, 82.9 per cent of Bul. 606] Control of Mealy Plum Aphid 19 TABLE 8 Fall Versus Spring Application with and without Nicotine ; and Fall and Spring Combination Location and material Ranch 1 Light oil, 2.0 per cent. Check No. 1. Check No. 2. Ranch 2 Whale-oil soapf. Check No. 1. Check No. 2. Ranch 3 Whale-oil soapf. Check No. 1. Check No. 2. Ranch 4 Fish-oil soapt . Check No. 1. Nicotine included" Yes... No. Yes. No. Yes.... No. Season 'Fall Spring ^Fall and spring r Fall Spring Fall and spring Fall I Spring [Fall and spring [Fall J Spring i Fall and spring Fall Spring ^Fall and spring 'Fall Spring Fall and spring Spring . Spring . Number of trees 30 29 32 25 28 29 34 48 36 32 35 48 41 36 48 90 17 17 26 18 24 25 15 12 22 42 16 Per cent of trees Heavily infested 80.0 10.3 6.3 88.0 32.2 24.2 100. p 77.2 52.8 15.7 8.6 37.5 34.1 19.4 35.5 60.0 94.1 11.7 15.3 16.7 16.7 12.0 20.0 91.7 4.5 30.9 68.2 Lightly infested 13.3 31.1 43.7 12.0 46.4 27.6 0.0 14.5 16.7 34.3 34.3 18.7 31.8 41.7 33.3 22.2 5.9 64.7 73.1 72.2 83.3 76.0 60.0 8.3 36.4 21.4 18.8 Clean 6.7 58.6 50.0 0.0 21.4 48.2 0.0 8.3 30.5 50.0 57.1 43.8 34.1 38.9 31 2 17.8 0.0 23.6 11.4 11.1 0.0 12.0 20.0 0.0 59.1 47.7 12.4 * Yi pint nicotine sulfate per 100 gallons. t 3 gallons per 100 gallons. % 4 lbs. per 100 gallons. the eggs failed to hatch. The high efficiency of coal-tar distillate and of ortho-dinitro-cresol led to the testing of coal-tar distillates in the field as reported later. Fall Applications Compared and Combined with Spring Applications, with and without Nicotine. — During the fall of 1930 and spring of 1931, field tests were conducted to compare the efficiency of fall and spring applications, and combinations of these two. In each case the sprays were 20 University of California — Experiment Station applied with and without the addition of nicotine sulfate. The results given in table 8 were obtained. Ranch 3, because of several factors, such as uneven distribution of aphids in the orchard, and unusual activity of predators in the spring, did not yield consistent results. The best results were obtained on ranch 1, which is diagramed in figure 7. For a simple 3 ' 4 ' 5 ' 6 ' s Fig. 7. — Fall and spring applications alone and combined ; with and without nicotine; map of ranch 1, table 8. Plot 1, not sprayed; plot 2, oil and nicotine in the fall; plot 3, oil and nicotine in the fall and spring; plot 4, oil and nicotine in the spring; plot 5, oil alone in the spring; plot 6, oil alone, fall and spring ; plot 7, oil alone in the fall ; plot 8, not sprayed. Legend as in figure 5. reading of the results in this case, in table 9 the control has been ex- pressed in percentage reduction per penny spent for spray materials. The data in table 8 indicate that a spring application is far more ef- fective than a fall application. The reason for this, as indicated earlier, is the fact that continued migration in the fall results in reinf estation of the trees after spraying. In nearly every case, the addition of nicotine sulfate to oil or soap yielded a measurable increase in control. By re- ferring to the figures in table 9 it can readily be seen that the nicotine increased the control slightly more rapidly than it increased the cost of the spray mixture. Early Field Tests with Coal-Tar Distillates. — In view of the high ovicidal efficiency of coal-tar distillate in laboratory tests conducted in 1930-31, this material was tested in the field during the winter of 1931- 32. In most cases the efficiency of coal-tar distillates was tested in com- parison with a spring application of oil and nicotine, which was the most effective spray developed up to that time. The data in table 10 were ob- tained. These figures indicate that coal-tar distillate is far more efficient Bul. 606] Control of Mealy Plum Aphid 21 TABLE 9 Fall and Spring Applications of Light Petroleum Oil, with and without nicotine* Season Nicotine added Gallons per tree Cost of materials per tree.f cents Reduction^ in percentage of heavily infested trees Reduction in percentage of heavily infested trees per penny of spray material Fall Spring Fall and spring /No \Yes /No \Yes , [No < Yes 2.50 2.50 1.76 1.64 / 2.50 1 I 1 76/ / 2.50 \ I 1.64/ 2.5 3.3 1.8 2.2 4.3 5.4 0.6 8.6 56.4 78.3 64.4 82.3 0.2 2.6 31.3 / 35.6 15.0 15.2 * Based on the experiments reported in table 8. t Computed from the arbitrarily chosen values of light oil $0.50 a gallon, nicotine sulfate $5.00 a gallon. J Computed as the difference between the percentage of trees heavily infested in the sprayed area, and the average percentage of trees heavily infested in the two check areas. TABLE 10 Coal-Tar Distillate Compared with Petroleum Oil and Nicotine, Placer and San Joaquin Counties, Winter of 1931-32 Material Dilution, per cent Number of trees Per cent of trees infested Ranch Heavily Lightly Clean Coal-tar distillate 3.0 2.0 0.125 48 \ 122 76 2.1 23.8 85.5 2.1 19.7 6.6 95.8 1 Petroleum oil 56.5 Nicotine sulfate Not sprayed 7.9 2 /Coal-tar distillate 3.0 71 22 31.1 100.0 62.3 0.0 6.6 \Not sprayed 0.0 Coal-tar distillate 3.0 2.0 0.125 43 7.0 74.4 4.7 11.6 88.3 3 Petroleum oil 14.0 Nicotine sulfate Coal-tar distillate 3.0 2.0 0.125 21 \ 46 4.8 76.1 14.3 4.3 80.9 4 Petroleum oil 19.6 Nicotine sulfate 22 University of California — Experiment Station in controlling the aphid than an oil-and-nicotine spray. To illustrate the arrangement of the blocks and the location of the infestations, ranches 1 and 3 are represented diagrammatically in figure 8. In these tests the varieties Grand Duke, Diamond, President, and French were sprayed. Three of the ranches were located in Placer County and one in San Joaquin County. No traces of spray injury appeared on any of the trees. <& 4 O/L AND / NOT /COAL TAN NOT / O/L AND N/COT/N£ /SPNAYCD/D/ST/LLATf / SPNArfD/ N/COT/N£ B ^^J^^X^^X^X^X^X^^^X^ < 6$£> &® COAL TAN <&<& D/ST/LLATC 4$k@$> f£4^ O/L AND &<&& N/COT/NE Fig. 8. — Comparison of coal-tar distillate with oil and nicotine in the spring: A, ranch 1, table 10 ; B, ranch 3, table 10. Legend as in figure 5. On the contrary, the sprayed trees were more vigorous than the check trees. This is believed to have been due to the absence of aphids rather than to the direct physiological action of the spray. The following winter (1932-33) coal-tar distillate was tested at sev- eral concentrations to determine, if possible, the lowest efficient strength. Field plots were laid out using coal-tar distillate at 0.6, 1.2, 1.5, 1.8, 2.4, and 3.0 per cent of actual coal tar. The results obtained on French prunes in San Joaquin County are given in table 11. The applications at 1.5 and 3.0 per cent, which were located in Placer County, yielded no data since both sprayed areas and checks were free from aphids the following Bul. 606] Control of Mealy Plum Aphid 23 spring. In table 11, the plots are listed in the order in which they were located in the field ; thus the 2.4 per cent application was located between check area No. 1 and the 1.2 per cent application. The arrangement of the plots and the results of the & <&4£ A/or SP/?AY£D L8% test are indicated for ranch 2 in figure 9. The data for ranches 1 and 2 are not directly compar- able owing to variation in fac- tors in each case. The trees on ranch 1 were old, and too close together, which resulted in high, umbrella-shaped tops, which were difficult to spray thor- oughly. The data indicate that under certain conditions a good con- trol may be obtained with a con- centration as low as 0.6 per cent. On ranch 1, doubling the con- centration of tar increased the clean trees by 34.4 per cent ; while on ranch 2, tripling the tar concen- tration increased the clean trees by only 8.9 per cent. This difference is due in part to the higher percentage of clean trees resulting from the lesser application on ranch 2. TABLE 11 Field Tests of Various Concentrations of Coal-Tar Distillate on French Prunes, San Joaquin County, 1932-33 0.6% # #> #> # 4£ # A/or 3PRAr£D Fig. 9. — Coal-tar distillate; ranch 2, table 10. Legend as in figure 5. Ranch Dilution, per cent Not sprayed No. 1. 2.4 1.2 Not sprayed No. 2. Not sprayed No. 1. 0.6 1.8 Not sprayed No. 2. Per cent of trees infested Heavily 100.0 11.4 27.5 66.7 69.2 2.9 5.3 87.1 Lightly 0.0 45.7 40.0 33.3 26.2 29.4 21.0 12.9 Clean 0.0 42.9 32.0 0.0 4.6 67.7 73.7 0.0 Tests with Ten Proprietary Coal-Tar Sprays. — During the winter of 1933-34, nine domestic and one imported proprietary tar sprays were studied. Laboratory tests on cut twigs, as described earlier, were per- 24 University of California — Experiment Station TABLE 12 Toxicity of Tar Sprays on Plum-Aphid Eggs, Laboratory Tests, 1934 January 14 February 14 Brand 1 per cent tar 4 per cent tar 1 per cent tar 4 per cent tar No. Number of eggs used Per cent of eggs not hatched Number of eggs used Per cent of eggs not hatched Number of eggs used Per cent of eggs not hatched Number of eggs used Per cent of eggs not hatched I 158 141 176 211 301 276 109 150 199 226 194.7 621* 83.5 95.0 83.0 99.5 100.0 90.9 99.1 91.3 98.0 98.2 93.9 38.4 117 121 177 155 165 143 161 176 225 97 153.7 l,238f 100.0 100.0 98.9 99.4 99.3 99.3 99.3 98.9 99.6 96.9 99.2 40.3 279 195 102 87 214 102 133 229 82 101 152.4 86.7 99.0 60.8 96.6 98 6 90.2 99.2 92.8 97.6 75.2 89.7 183 158 167 272 342 228 105 230 210 88 198.3 100.0 II 100.0 Ill 100.0 IV 100.0 V 100.0 VI 100.0 VII 100.0 VIII 100.0 IX 100.0 X 97.7 Average 99.8 Check * Check sprayed with water, f Check not sprayed with water. TABLE 13 Relation of Emulsification to Toxicity; Per Cent Kill of Viable Eggs* with One Per Cent Tar Distillate; Laboratory Tests, 1934 Type of emulsion January 14 February 14 Average Group average Miscible 72.6 91.7 84.9 85.5 97.0 99.1 100.0 ] 98.5 96.6 77.9 98.3 83.7 88.0 58.8 94.3 97.6 98.6 96.0 75.3 95.0 84.3 86.8 77.9 96.7 98.8 98.5 96.3 83.9 Paste t \ 97.6 0>.g 03 « 03 1-H CO CO CO >o o £ oo CO e» CO ■* CO o t^ ^H > CD CO t^ t^ CO es «5 CO «! ■3 a 1 d » "" 2?d 3 diJ 8 1— ( ^H OS CO C» 05 >-H o o CO 1-H CO CO r-- CO 1-H t^ CO ^ O lH«*5 d « ° >. >-i a +j d d 03 d o >-ho o» o ^o o>o NO t»< O oc co e M o O ts 0> i-h O COO O) O (NO t-- O »-i© oc "S c ~H O G CO CO r~ CO *-H t^ CO ■•f '3 a* 03 d o 0) j>> d 53 CO 111 coco oeo t»o cn o i-h io coo oo c w oo >o oo r-co h* co t>- i-h oo o i-h c CO c 1-H G 1-H O ■#H eo c« 1-H CO CO c<« 1-H CO H»< »o o 3 CO 43 I— I d o ^ o >> 5 Ph j>> (h 03 a '> tO N i-h r- Si O OiO t— »C OO CNC 1-H C t» o o3 lO r* *-i ■*■ * Oil— CNJt~ IQ 00 oo ooc CO c •»»« o O) CD c 1-H O 03 n ^ ^ F— 1-H h 3 PI I CO CO CO CO Hf CO CM CC •»J< ^1 COCO COCO COCO COCO COCO N M co o: CO CC CO CO z, a a> .3 M fl 8-S «8 55 o> > d a o c o t~ a oc o oo t^ c o c 1-H OC o c m t~ d 9 T3 o> io c o cc (M t^ O >fl «— i o o « t- t^ o c CM »-H $ o t~ o c: OS «N o »- •^f »o ■* O iff CO h* o *5 T-t 1-H 1-H O $ a> "9 u 03 m 9 >, 0> ■p O »H o cz i— l *— o w: CO »- O O IS t> o c oo >o ■** O Oi o c t^ PH o c 00 »H - ■- oo oc o ": oo e1 d t-l o> o3 o d d 03 •-5 (h Oi V 03 o CJ- O K o — o t^ o os o c CO >rt o c OO oo US C O e«: o •— o C. o oc o c •^< CC o «: OO lC 3 > > > >— > X! ^ ^ ^ M M ^ ^ ^ ^t e e t c c c hH a c e a 3 s CL ■ i - X hH ^ 1— 1 X 1-H X tXj -£ X HH c R > > G > G > G P G X G Bul. 606] Control of Mealy Plum Aphid 27 in a checkerboard pattern, as shown in figure 10. Alternating with each block of 16 sprayed trees were staggered groups of 8 unsprayed trees. In computing a check value for each sprayed area, all unsprayed trees within a radius of two trees of the sprayed area were counted. This would allow 24 unsprayed trees for each sprayed block of 16 trees ; that is, 8 check trees on each of two sides and 4 check trees on each of two ends of the sprayed area. The sprays applied in Jan- uary, 1935, were applied in groups of three contiguous blocks with comparable checks between each grouping of three sprays. Thus four check areas were established. The data from these tests are given in table 14. The differ- ence in efficiency between mis- cible and paste emulsions dem- onstrated in the laboratory tests (table 13), were further observable in the field tests. For miscible types the increase in clean trees was 68.1, 63.6, 4.5, -20.4, and 31.0 per cent— an average of 29.4 per cent; for paste types it was 72.3, 73.6, 63.2, and 57.8 per cent— an average of 66.7 per cent. 4°/ w (@> <§b<££> cgb <§£> <£§!> <& <& @> <@> #» 4gt <& #> <&> ^ <& <@> c//fc/r/ #,4ffi#>#>#> 4JL# ### cgb # (gi # (§*> # <@> ##» # <££> Fig. 11. — Coal-tar distillate at various con- centrations; material A, table 15. Legend as in figure 5. Tests to Determine the Minimum Effective Concentration of Tar Dis- tillate. — During January, 1935, two proprietary, paste-type emulsions were tested at various dilutions. The arrangement of the plots and checks # is illustrated, for material A, in figure 11. The other tests were arranged similarly. The data in table 15 were obtained. When the data obtained in the three tests are averaged, the increase in percentage of clean trees for the various concentrations is as follows : 0.5 per cent tar, 24.2 ; 1.0 per cent tar, 61.8 ; 1.5 per cent tar, 71.3 ; 2.0 per cent tar, 71.0 ; 4.0 per cent tar, 83.9. These data are represented graphically in figure 12. 28 University of California — Experiment Station TABLE 15 Concentration and Efficiency of Coal-Tar Distillates, January, 1935 Material Number of trees used Per cent of trees infested Increase in per cent and location Dilution,* per cent Heavily Lightly Clean of trees clean < < [0.5 37 27 33 33 23 34 16 27 32 28 40 40 32 54 44 28 69 129 24 0.0 51.9 0.0 0.0 73.9 0.0 0.0 100.0 87.5 100.0 40.0 27.5 100.0 31.5 13.6 100.0 4.3 2.3 87.5 56.8 40.7 45.5 21.2 17.4 29.4 18.7 0.0 12.5 0.0 60.0 50.0 0.0 55.5 56.8 0.0 13.1 10.9 8.3 43.2 7.4 54.5 78.8 8.7 70.6 81.3 0.0 0.0 0.0 0.0 22.5 0.0 13.0 29.6 0.0 82.6 86.8 4.2 35.8 Check 1 A, 1.0 47.1 San Joaquin 1.5 70.1 County Check 2 2.0 61.9 4.0 81.3 ^Check3 '0.5 Check 1 12. 5f B,t San Joaquin 1.0 1.5 60. Of 72. 5t County Check 2 2.0 68. 5t 4.0 86. 4f B, , Check 3 1.0 78.4 Placer County 1 2.0 Check 82.6 * Percentage of actual coal-tar distillate in the finished spray mixture. t In order to obtain contrasting data, both lightly infested and clean trees were computed into the percentage of clean trees in the last column for ranch B. 0.5 f.O /.5 2.0 2.5 3.0 3.5 4.0 CONCCNTRAT/ON OrTARD/Sr/UAr£ 9 P£R C£NT Fig. 12. — The relation of the concentration of coal-tar distillate to efficiency. Averaged data from table 15. Bul. 606] Control of Mealy Plum Aphid 29 DISCUSSION AND CONCLUSIONS The most effective method of controlling the mealy plum aphid consists of spraying with coal-tar distillate in the dormant or winter period, to kill the eggs. The tar distillate, as has been shown by many investigators, should consist of the high-boiling, neutral hydrocarbons. The low effi- ciency of the low-boiling fractions is indicated by the relative failure of brand III, table 12. Practically all of the proprietary tar sprays on the market consist of high-boiling fractions. Emulsification. — Greater efficiency was obtained from tar sprays pre- pared as paste-type emulsions than from miscible types. This may be due to a greater deposit of tar distillate from the former type. Concentration. — The concentration of tar in the finished spray that is necessary to yield a satisfactory control is determined by the density of aphid population, by the efficiency of the spray rig and spray crew, and by the amount of spray applied per tree. In the second application of material B, table 15, a good control was obtained with a concentration of only 1 per cent tar. In this case, the trees were small and were sprayed with rods using a pressure of 250 pounds. The spray crew consisted of experienced men. The aphid population was moderate, since 4 per cent of the check trees remained free from aphids. In the other two tests in table 15, a concentration of 1 per cent failed to yield a satisfactory con- trol. In these tests large trees were used, spraying was done with guns, and owing to several factors, including wind, a poorer coverage was ob- tained. The graph (fig. 11), indicates that, on the average, as the concen- tration of tar was increased to l 1 /^ per cent, the control increased rapidly, but as the concentration of tar was increased from 1% to 4 per cent the increase in the control was much slower. The concentration of tar which should be used, therefore, should be 1% or 2 per cent, according to density of aphid population and efficiency and thoroughness of the ap- plication. The figures on concentration of tar used throughout this paper refer to the actual percentage of coal-tar distillate in the finished spray mix- ture in the spray tank. The proprietary tar sprays, as marketed, contain from 50 to 83 per cent of tar distillate. The concentration of tar is stated on the manufacturer's label on the drum. The necessary dilution is com- puted by dividing 100 by the percentage on the label and multiplying by the concentration desired in the finished spray. Thus, to make a 1% per cent emulsion from a stock containing 83 per cent of tar distillate : 100-^-83 = 1.2; 1.2 X 1.5 = 1.8 gallons of stock per 100 gallons of water. 30 University of California — Experiment Station Some proprietary brands of tar sprays contain an appreciable amount of petroleum oils. As has been shown (tables 2 and 6) these oils are but slightly toxic to mealy-plum-aphid eggs. Dilutions, therefore, should be based on the coal-tar content and not on the combined coal-tar and pe- troleum-oil content. Injury to the Trees by Coal-Tar Distillate. — During the course of this investigation, no injury to the trees appeared in any of the tests. Since its introduction, a considerable amount of tar spray has been used com- mercially and no injury has developed except in El Dorado County, where severe injury resulted from tar applications. The reason for this localized condition is unknown. It is suggested that tar sprays be used sparingly the first year in new localities until it is definitely established that no injury will occur. Tar sprays have been used in the Sacramento, San Joaquin, Santa Clara, and Sonoma valleys without injury to the trees. In addition to several varieties of blue plums and prunes, tar sprays were applied at 3 per cent to Black Tartarian cherries ; Elberta, Levi, and Triumph peaches ; Climax and Formosa plums ; and Bartlett pears, in Placer County. No injury appeared in any case. Tar sprays are exceedingly toxic to trees in leaf. They cannot be ap- plied with safety after the buds have started to swell in the spring, and should not be applied before the first heavy rain in the fall. All of the ten brands discussed earlier were applied to French prune at dilutions of 3 to 4 per cent, on December 20, January 6, January 26, and February 27. No injury could be discovered the following spring as a result of any of these applications. In general, however, January is probably the safest month in which to apply tar sprays. Stimulation of Trees by Tar Sprays. — Throughout the course of the spray tests reported in this paper, trees were observed for stimulation. Stimulation was considered from two standpoints : (1) earlier blooming, and (2) increased growth throughout the ensuing season. Several of the proprietary brands contained various amounts of petroleum oils. These brands caused earlier blooming in proportion to the amount of petroleum oil which they contained. Those types which contained no petroleum oil did not advance the time of blooming. Applications of tar sprays are generally followed by an obvious in- crease of growth of the tree during the subsequent season, as shown in figure 3. This increase in growth has been measured and is recorded on page 7, wherein six sprayed blocks showed a definite increase in length and weight of new wood, with an average increase of 15.9 per cent in length and 66.7 per cent in weight over check trees. The question arises Bul. 606] Control of Mealy Plum Aphid 31 as to whether this increase in growth is a direct physiological stimulation by the tar, or whether it is simply normal growth which contrasts very favorably with the growth of unsprayed, heavily infested, check trees. The coefficient of correlation between the percentage increase in length of new growth and the percentage of clean trees in the sprayed blocks is 0.22. Although this correlation is low, the writer believes that the in- crease in growth of sprayed trees is due largely to the absence of aphids, rather than to a direct stimulation of the trees by the tar distillate. Effect of Tar Sprays on Men. — Tar sprays are caustic to the skin of men and horses, and precautions must be taken to avoid injury. The causticity varies with conditions and individuals. In-general, blondes are more susceptible to injury than brunettes, and hence, brunettes should comprise the spray crew whenever possible. Again, tar sprays produce a severer irritation in sunlight than in shade, and hence night spraying or spraying on overcast days will avoid some injury. Sprayers should protect all parts of their bodies from the spray. Spray coats, gloves, and hats should be worn, and the face should be protected with a sheet of moisture-proof cellophane wedged under the hat band. As the spray covers the cellophane and clouds the vision, additional clear space can be pulled out of the hat. Usually a fresh piece of cellophane will be needed for each tank of material. Attempts to protect the face with creams and greases have, in the case of the writer, proved unsuccessful. Some sprayers, however, claim that protection is obtained by the use of such materials. There is a tendency among sprayers to discard protective clothing and masks on the basis that they hamper their vision and action, or that such protection is unmanly. This is a serious mistake. The possible injury from tar sprays should not be taken lightly. The writer has seen several men severely burned, and one required hospitalization. As has been stated earlier, men vary in their susceptibility to tar injury. The fact that one man experiences no difficulty is no clue to the susceptibility of his co-workers. Each man should determine to his own satisfaction whether or not he is susceptible. It should be borne in mind that the burning sensation does not arise immediately after an application of tar to the skin, but rather will reach a maximum from 2 to 4 hours later. Deciding Whether to Spray or Not to Spray. — The question as to whether or not spraying is warranted is a most difficult one in certain areas, because of the spotty distribution and fluctuation in aphid occur- rence from year to year. In some sections, particularly in San Joaquin and Placer counties, the aphids are injurious almost every year. During the past nine years, the aphids have been injurious in San Joaquin 32 University of California — Experiment Station County every year, and in Placer County every year but one, when they were of minor importance only. In such areas, spraying each year is recommended. In the Santa Clara and Sonoma valleys, the occurrence of injurious numbers of aphids is less frequent. Sections of these valleys vary from complete absence of aphids to a condition where injurious infestations occur four years in every five. In such cases each grower will need to decide whether or not to spray, on the evidence pertinent to his own orchard. In making this decision, the following points should be kept in mind. 1. There is no relation between the infestation of the past season and that of next season. The fact that aphids were present in the orchard one year does not indicate that they will be present the following year, since they do not complete their yearly cycle on the trees. If, however, the grower has information on the amount of aphid infestation that has occurred during the past five or ten years, he may estimate the chances of the next year on that basis. 2. In general, it is impossible to determine the question of spraying by an inspection of the orchard in midwinter and a count of eggs. In cases of abnormal egg deposition, as shown in figure 1, eggs may easily be found, and if so, the trees should be sprayed. Under favorable condi- tions, 10 to 20 eggs are sufficient to give rise to a heavy aphid infestation by summer. The difficulty of finding the eggs when they are present in such limited numbers renders a search for them futile. A much more feasible survey can be made in the fall, of the fall mi- grants and sexual females. Such an inspection is best made when 80 to 90 per cent of the leaves have fallen from the trees. As the leaves yellow preparatory to falling, the aphids leave them and move to the leaves which still remain green. In this way, they are ultimately concentrated on the last few leaves. If three or four aphids can be found on each tree at this time, control measures are warranted. 3. The best trees to inspect for aphids are those which are located in a sheltered area, since the returning migrants "settle out" of the still air in such places and produce a much heavier infestation. Hill tops or crests of ridges are generally free from aphids, whereas swales are gen- erally infested. Heavy infestations usually occur near windbreaks, such as a row of tall eucalyptus trees or a large barn. Under certain condi- tions, therefore, it is necessary to spray only those trees which stand in sheltered situations. Other Advantages Derived from Tar Sprays. — While no particular study has been made of the action of tar sprays on other plum pests and diseases, evidence has come to hand that tar sprays reduce the incidence Bul. 606] Control of Mealy Plum Aphid 33 of brown apricot and soft brown scales, moss and lichens, and brown rot. It is believed that thorough spraying each year with 2 per cent coal-tar distillate will effectively control soft brown and brown apricot scales. Tar sprays have no appreciable effect on San Jose scale or leaf -roller eggs. Supplementary Methods of Control. — In the event that the tar spray has been neglected, the aphids can be checked to a considerable extent by a thorough application in the spring when about 5 per cent of the flowers are in bloom. The spray may consist of 2 per cent light summer-type petroleum oil plus % pint of nicotine sulfate per 100 gallons; or 3 gallons of fish-oil soap plus % pint of nicotine sulfate for 100 gallons of water. Either of these mixtures can be used shortly after the blooming period, but best results are obtained earlier. In some sections, as in the Hollister district, where the aphid infestation is very light, infested limbs are marked and sprayed individually or dusted with a strong nico- tine dust. SUMMARY The aphid spends the spring and early summer on the plum, and the fall . on the cattail and reed grass. A return migration occurs in the fall, and the overwintering eggs are laid on the trees. The injury produced by this pest consists of curling and stunting of foliage, lessened growth of new wood, soiling of the bloom of plums, and splitting of the fruit. Elimination of secondary hosts was found to be impractical for areas less than 30 miles in diameter. Fall defoliation with leaf -killing sprays was found to be undepend- able and dangerous to the tree. Tests of single sprays of petroleum oil or soap, with or without nico- tine, during the fall, winter, and spring failed to produce a satisfactory control. Tests of multiple applications of petroleum oil or soap indicated that as many as three applications fail to give a complete kill. Earlier applications of spring sprays gave better results than later spring sprays ; and spring sprays showed greater efficiency than fall sprays. The addition of nicotine increased the control to such an extent as to justify its use on a cost basis. Laboratory tests of petroleum oil showed a low kill of eggs, as was likewise the case in the field. Viscosity and unsulfonatable residue were not related to toxicity. Miscible petroleum-oil emulsions were not more toxic than paste types. An increase of petroleum oil in the diluted spray resulted in an increase in kill of eggs. Of several materials added to petroleum oil, dinitro-ortho-cresol, a coal-tar product, gave by far the 34 University of California — Experiment Station greatest increase in toxicity. A laboratory test of coal-tar distillate emulsion gave a high kill of eggs. Field tests of winter applications of coal-tar distillate in comparison to spring applications of oil and nicotine demonstrated a greater effi- ciency for coal-tar distillate. Tests in the field during the dormant period with five miscible and five paste-type coal-tar distillate emulsions indicated that the paste-type is more efficient. Laboratory tests corroborated this fact. Field tests of tar at five concentrations ranging from 0.5 to 4.0 per cent indicated that 1.5 per cent yields a good control when very efficient spraying is done and that 2.0 per cent is necessary for mediocre spraying. No injury has been done to the trees by tar sprays in any of the experi- ments reported herein. Increased tree growth follows an application of tar sprays, believed to be the result of absence of aphids. Tar sprays are caustic to men, and sprayers must protect themselves. A method of deciding whether or not to spray is discussed. ACKNOWLEDGMENTS During the course of this investigation the writer received valuable assistance from Mr. H. E. Butler and Mr. W. D. Bethell of the Penryn Fruit Company, as well as Mr. A. Swank, Mr. C. D. Chase, and Mr. Barton. These men loaned portions of their orchards and assisted in the application of the sprays. Mr. Gene Serr, Assistant County Agent, San Joaquin County, has been helpful in many ways. A number of spray- manufacturing companies have cooperated by furnishing desired ma- terials. To all of these men and organizations the writer extends his sincere appreciation. llm-5,'37