«^ •%!* f|^ f^ f^ LEASE HANDLE WITH CARE University of Connecticut Libraries 3 ^153 Dl^a^7S'^ i GAYLORD RG Digitized by the Internet Archive in 2011 with funding from LYRASIS members and Sloan Foundation http://www.archive.org/details/tobaccosubstatioOOswan Bulletin 350 S April, 1933 r^^^3S^ TOBACCO SUBSTATION AT WINDSOR REPORT FOR 1932 T. R. SWANBACK, O. E. STREET AND P. J. ANDERSON Glnntt^rttrut CONNECTICUT AGRICULTURAL EXPERIMENT STATION BOARD OF CONTROL His Excellency, Governor Wilbur L. Cross, ex-officio, President Elijah Rogers, Vice-President Southington George A. Hopson, Secretary Mount Carmel William L. Slate, Director and Treasurer New Haven Joseph W. Alsop Avon Edward C. Schneider Middletown S. McLean Buckingham Watertown Charles G. Morris Newtown STAFF Administration. William L. Slate, B.Sc, Director and Treasurer. Miss L. M. Brautlecht, Bookkeeper and Librarian. Miss Dorothy Amrine, B. Litt, Editor. G. E. Graham, In Charge of Buildings and Grounds. Analytical E. M. Bailey, Ph.D., C/jemi's/ in C/iar««. Chemistry. C. E. Shepard Owen L. Nolan Harry J. Fisher, Ph.D. \ Assistant Chemists. W. T. Mathis David C. Walden, B.S. Frank C. Sheldon, Laboratory Assistant. V. L. Churchill, Sampling Agent. Mrs. a. B. Vosburgh, Secretary. Biochemistry. H. B. Vickery, Ph.D., Biochemist in Charge. Lafayette B. Mendel, Ph.D., Research Associate (Yale University). George W. Pucher, Ph.D., Assistant Biochemist. Botany. G. P. Clinton, Sc.D., Botanist in Charge. E. M. Stoddard, B.S., Pomologist. Miss Florence A. McCormick, Ph.D., Pathologist. A. A. Dunlap, Ph.D., Assistant Mycologist. A. D. McDonnell, General Assistant. Mrs. W. W. Kelsey, Secretary. Entomology. W. E. Brixton, Ph.D., D.Sc, Entomologist in Charge, State Ento- mologist. B. H. Walden. B.Agr. 1 M. p. Zappe, B.S. Philip Garman, Ph.D. [Assistant Entomologists. Roger B. Friend, Ph.D. | Neely Turner, M.A. J John T. Ashworth, Deputy in Charge of Gipsy Moth Control. R. C. BoTSFORD, Deputy in Charge of Mosquito Elimination. J. P. Johnson, B.S., Deputy in Charge of Asiatic and Japanese Beetle Quarantines. Mrs. Gladys Brooke, B.A., Secretary. Forestry. Walter O. Filley, Forester in Charge. H. W. HicocK, M.F., Assistant Forester. J. E. Riley, Jr., M.F., In Charge of Blister Rust Control. Miss Pauline A. Merchant, Secretary. Plant Breeding. Donald F. Jones, Sc.D., Geneticist in Charge. W. Ralph Singleton, Sc.D., Assistant Geneticist. Lawrence C. Curtis, B.S., Assistant. Soils. M. F. Morgan, M.S., Agronomist in Charge. H. G. M. Jacobson, M.S., Assistant Agronomist. Herbert A. Lunt, Ph.D., Assistant in Forest Soils. D wight B. Downs, General Assistant. Tobacco Substation Paul J. Anderson, Ph.D., Pathologist in Charge. at Windsor. T. R. Swanback, M.S., Agronomist. O. E. Street, M.S., Plant Physiologist. Miss Dorothy Lenard, Secretary. Press of The Wilso)! H. Lee Co., Orange, Conn. CONTENTS Page Tobacco Projects 466 How Much Magnesia Should Be Applied To Tobacco Land .... 466 Plan of the experiment 467 Effect on yield and grading 467 Effect on soil reaction 468 Effect on character of the ash 469 Effect on chemical composition of the leaves 471 Conclusions and recommendations 473 The Relation of Calcium to the Growth of Tobacco 473 Percentage of calcium in the leaf 474 The reciprocal relation of calcium and other bases 475 The effect of various compounds of calcium 476 Effect on yield .-...-. 476 Literature cited 478 Further Experiments With Nitrophoska 478 Broadleaf Fertilizer Experiments 479 Sources of nitrogen 480 Quantity of phosphorus 481 Conservation of Plant Nutrients by Cover Crops 482 Quantity of water leached 483 Temporary withdrawal of nitrogen by the cover crop 483 Effect of cover crop on the leaching of nutrients 485 Lysimeter results at other stations 487 Tobacco Insects in 1932 488 Prevalence of various species 488 Potato flea beetle 489 Tobacco budworm 49 1 Tarnished plant bug 51-93 Distribution of wireworm larvae in tobacco soil 494 Insecticide tests against wireworms 497 Control of thrips 498 Comparative Studies of Fuels for Curing in 1932 499 Shade Curing Experiments in 1932 502 Preliminary runs 502 Humidified versus non-humidified sheds 503 Summary 506 \y- (463) TOBACCO SUBSTATION AT WINDSOR REPORT FOR 1932 T. R. SwANBACK, O. E. Street and P. J. Anderson This bulletin contains reports of progress on a number of the lines of investigation underway at the Tobacco Substation in Wind- sor. None of these are final reports, but some results have been obtained that are of sufficient interest to growers and others work- ing along these same lines, to warrant publication at this time. No mention is made here of the majority of the projects, a complete list of which will be found on page 466. The scope of the fertilizer experiments has been considerably changed, so that most of the tests in 1932 were in the first year of long-time experiments and therefore warrant no report at this time. The potash tests for the most part were discontinued with the season of 1931. New series of tests dealing with nitrogen fertilizer materials and rates of application of nitrogen occupy the greater part of the experimental farm at Windsor. In previous fertilizer experiments at Windsor, only the Havana Seed type of tobacco was used. In the new series of fertilizer experiments both Havana Seed and Broadleaf types are under test here. The Broadleaf fertilizer tests on the Frank Roberts farm, Silver Lane, were also continued under the direction of J. S. Owens, Extension Agronomist. Manure experiments on shade tobacco were continued in cooperation with Tudor Holcomb on his plantation in West Granby. A new series of curing experiments on Shade tobacco was begun this year in cooperation with the Gershel-Kaffenburgh Tobacco Company on their plantation in East Hartford. The writers take this opportunity to express their appreciation of the excellent cooperation of these growers. Since the appearance of our report for 1931, two other bulletins dealing with tobacco have been published by Connecticut investi- gators. One is "A History of Tobacco Production in New England," by C. I. Hendrickson, Bulletin 174 of the Storrs Experiment Station, which should be read by every tobacco grower. The other is "Chemical Investigations of the Tobacco Plant, III, Tobacco Seed," Bulletin 339 of the Agricultural Station, New (465) 466 Connecticut Experiment Station Bulletin 350 Haven. This bulletin is of special interest to the scientific inves- tigator. TOBACCO PROJECTS 1. Fertilizer experiments — comparison of carriers and rates of application of nitrogen, phosphorus, potassium and magnesium. 2. Field tests with farm manure. 4. Tobacco nutrition studies — the roles of magnesium and calcium and rates of nitrification of different fertilizers. 5. Improvement of Havana Seed tobacco with especial reference to root- rot resistance. (With U. S. Dept. Agr.) 7. Improvement of Cuban Shade strains. 8. Cover crops for tobacco fields. 9. Brown rootrot, nature and causes. II. Soil reaction in relation to tobacco culture. 13. Preservative treatment of shade tent poles. (With Department of Forestry.) 17. Factors that influence curing. 19. Investigation of tobacco diseases. 20. Insects of tobacco. (With Department of Entomology.) 23. Metabolism as correlated with the stages of growth of tobacco. HOW MUCH MAGNESIA SHOULD BE APPLIED TO TOBACCO LAND? The importance of magnesia, not only for the proper growth of tobacco but more particularly for improvement of combustion of the cigar has been fully demonstrated at this Station.^ It M-as found that the percentage of magnesia in the leaf could be readily increased by application of magnesia-containing materials to the soil, in the fertilizer mixture or alone. Although the organic fer- tilizer materials, such as cottonseed meal, linseed meal, or castor pomace, contain some magnesia, the quantity thus supplied is usu- ally not sufficient to produce optimum improvement in burn.^ Larger amounts may be applied in cottonhull ash or sulfate of potash-magnesia. The most economical material for this purpose is magnesian lime or limestone containing a high percentage of MgO. When the proper amount of magnesia is present in the leaf, com- bustion is more complete, resulting in a vi^hiter ash, closer burn (narrow coal band), and better taste and aroma. If, however, the magnesia is increased too much, the ash falls off the cigar in flakes, and in that respect the burn is not desirable. The optimum amount of magnesia in the leaf therefore is that percentage which is suffi- cient to give a light-colored, close burning ash, but is not enough to cause "flaking." It was tentatively concluded from experiments, chemical analyses, and burn tests, that this optimum percentage is obtained when the leaf contains about 2 per cent of MgO (calcu- lated on moisture free basis). * For a full discussion of experiments conducted and conclusions drawn, the reader is referred to our previous reports, particularly Bui. 326, pages 391-398, 1930, ^ "Burn" refers to the composite effect of bases in the combustion of a cigar, thus differing from "fire holding capacity," which is primarily influenced by the potash content of the leaf. Applying Magnesia to Tobacco Land 467 From a practical standpoint it was next necessary to determine how much magnesia should be applied to an acre of land to insure a magnesia content of about 2 per cent in the leaf. Obviously the required amount will depend somewhat on the character of the soil, its native magnesia content, and the season. Despite these variable factors, it seemed desirable to determine by field tests as nearly as possible the optimum quantity to apply and how often it should be applied to an average tobacco soil such as we have on the Station farm. An experiment was therefore begun in 1930 on a field that was then in grass and had not grown tobacco during the preceding 9 years. Plan of the Experiment The plan was to apply different quantities of magnesia to adjacent plots of tobacco ; then, by burn tests on the tobacco of each plot, to determine which had the optimum burn and to correlate this with the actual percentage of magnesia found in the leaf by chemical analyses. Magnesian lime with a MgO content of 30 per cent was used for this purpose (57 per cent CaO). It was also necessary to keep records of growth, yield, and sorting in order to see what effect each treatment had on these factors and to deter- mine soil reaction at intervals, since there was danger that such applications might neutralize the soil to such an extent as to favor black rootrot. There were 10 plots of one-twentieth acre size on which four treatments in duplicate were applied : 100, 200, 400, and 600 pounds of MgO to the acre. The other two plots received no lime and served as controls. The application of magnesian lime was broadcast 2 weeks before setting in 1930. No lime was applied in 1931 or 1932. A general fertilizer was applied equally to all plots each year. This supplied 200 pounds of nitrogen, 169 pounds of phosphoric acid and 200 pounds of potash to the acre. The composition of the fertilizer (for an acre) was as follows : Cottonseed meal 1500 lbs. Castor pomace 500 Nitrate of potash 300 Sulfate of potash 70 Calurea 106 Precipitated bone 300 Effect on Yield and Grading- Observations during the growing season in these three years have not shown significant differences in growth where different quan- tities of lime were used. The tobacco from each plot was weighed and sorted separately and the grade index computed according to Connecticut Experiment Station Bulletin 350 ,«o Connecticut r..^v>^' ^>""^- -- ,He scale usedinaU the WO.U at .HsS^^o^^^^ Lsian lime improved somewhat bot^y.eld a g^^^^ .^ ^,,1^ ^nd -dr,ri?r::::uto'hr:gSfica„t. Effect on Soil Eeaction Naturally, the application of ^^f.^^^^^^f ''Vte'r"act"oroV?h^^ atleastte4orarily,theac,dj^y of the ol^^ .^ ^^^^^ „,3 4.93_ SuSs'f^S ^tlXva^S. the three years are presented '" ^^ the danger of f^^'^X^^fH^'^^^^''''^^ kept below 6.0O pH (f ^^'^S 600 pounds MgO, kept the sod m hirfiest two applications, 4<» ana °"^r„piore these amounts cer- Sanger .one for 'hjee y-- Therefore ^^^^.^3^ p^^^^^^ Tring It within the danger .one. Applying Magnesia to Tobacco Land 469 Table 2. Reaction of Soil on Limed Plots, 1930-1933 (pH) Reaction at different times of sampling Quantity of MgO applied to an acre July 8. Aug. IS. April 16. May 10, Jan. IS. 1930 1930 1931 1932 1933 0 4.90 5-23 5-45 5-53 5-33 100 5.06 5-48 5.60 5-48 5-45 200 5-36 6.15 5-87 5-72 5-55 400 5-95 6.65 6.31 6.07 5-90 600 6.53 7-34 6.89 6.21 6.03 Seasonal fluctuation in soil reaction makes it difficult to follow the yearly trend of reaction in each plot, but by comparing each plot at any one time with the control, it is apparent that the plots with the higher applications are becoming relatively more acid each year and are approaching the reaction of the control. This is obvi- ously due to greater loss by leaching from the high lime plots of calcium and magnesium. Effect on Character of the Ash In order to observe the effect of each increase in magnesia application on the burn, cigars were made from the fermented leaves (seconds) of the crops of 1930, 1931, and 1932. Wrapper, binder and filler were from the same plot. These were smoked and notes made on the color and coherence of ash. On those made from the first crop (1930) after liming it was found that the ash was very light gray to white on all cigars from the limed plots, but usually too dark on the control plots. Each increase in amount of magnesia, however, made the ash more flaky. (Figure 129.) This flakiness made the ash quite unsatisfactory in the heaviest three applications. Where 100 pounds of magnesia were applied, however, both the ash color and its coherence were satisfactory. Of the different quantities tried, this then was the optimum for the first year. The effect of magnesia on the second crop after application was shown when the cigars of the crop of 1931 were tested. (Figure 130.) Ash from the no- magnesia plots was now quite dark and "muddy," the taste and aroma were poor. Even where 100 pounds magnesia had been applied, the ash was not light enough in color to be satisfactory. On the 200-pound plots, the color, taste and aroma were satisfactory. With higher applications the ash became too flaky, but color, taste and aroma were excellent. It appears then from this year's tests that 100 pounds of mag- nesia are not sufficient to carry its beneficial influence into the second year. Only the 200-pound application was now satisfac- tory. Below this, the ash was too dark; above this, it was too 470 Connecticut Experiment Station Bulletin 350 1.32% Magnesia (MgO) Applied per Acre (lbs.) 100 200 400 Magnesia Found in the Leaf 2.47% 3-13% 3-83% 600 4-59% Figure 129. Partly smoked cigars from crop of 1930. Magnesian lime applied in spring. Magnesia (MgO) Applied per Acre (lbs.) 100 200 400 Magnesia Found in the Leaf 1.26% 1-55% I -90 600 .86% 1.26% 1.55% I-90 1.96% Figure 130. Cigars from crop of 1931. Second year after liming. Applying Magnesia to Tobacco Land 471 flaky. However, since the 200-pound application was not satis- factory the first year, it appeared that in practice it would be best to apply a smaller dose (100 pounds or less) every year and not depend on a "carry-over" effect of larger applications. Figure 130 shows how the ash become lighter with increasing quantities of magnesia applied to the soil. Cigars of the third year (1932) after application had a darker colored ash all through than corresponding plots of preceding years but it was not so flaky. (Figure 131.) Only the highest two applications gave satisfactory ash color and they were not as nearly white as in preceding years. Magnesia (MgO) Applied per Acre (lbs.) 100 200 400 Magnesia Found in the Leaf .92% .92% 1.14% 600 .87% .92% .92% 1.14% . 1-53% Figure 131. Cigars from crop of 1932. Third year after liming. Effect on Gbemical Composition of the Leaves It was to be anticipated that each increase in quantity of mag- nesia applied to the soil would be reflected in an increased percent- age of magnesia in the leaf. Previous experience also lead us to expect at the same time a decrease in the percentage of the other principal bases, calcium and potassium. In order to determine the extent of these changes and correlate them with the observed ash characters, analyses were made of the crops of 1930, 1931 and 1932. The average percentage of the three bases in the three crops is presented in Table 3. The percentage of magnesia (1.32) in the control plot of 1930 was somewhat higher than anticipated, but 472 Connecticut Experiment Station Bulletin 350 may possibly be accounted for by the previous cropping and treat- ment of the field. With this percentage of magnesia, the ash was usually somewhat too dark. Each increase in applied magnesia resulted in a sharp increase in percentage in the leaf. The adverse effect of too much magnesia is seen in the extreme flakiness of the ash as shown in Figure 129. The next two years each showed a sharp reduction in the percentage of magnesia in the leaf. The differences between the control and the treated plots became less each year indicating a rapid leaching of magnesia from the soil. It is obvious that the optimum amount of magnesia can not be maintained in the soil by large applications at intervals of several years but in such a soil the grower must depend on smaller annual applications. Table 3. Effect of Magnesium Lime on The Percentage of Calcium, Magnesium and Potassium in The Leaf. Air-Dry Basis on Unfermented Leaves in 1930 and 1932, Fermented in 193 i Percentage found in leaves Pounds MgO K2O CaO MgO applied per acre 1930 193 1 1932 1930 1931 1932 1930 1931 1932 None 4.83 6.32 6.45 6.75 5.36 5.26 I 32 .86 .867 100 3.9H 5-29 5-73 6.22 6.01 5.22 2 47 1.26 .917 200 3-12 5.06 4.89 5-63 5-36 5-51 3 13 1-55 .917 400 3-09 5.i« 5-67 5.26 5-07 4-55 3 «3 1.90 1.136 600 2.40 4.69 4-94 4-95 4.72 4-«5 4 59 1.96 r-534 Can we now correlate the percentage of magnesia in the leaf with the burn characteristics? The optimum burn for the first year was on cigars with a magnesia content of 2.47 per cent but even so the ash was too flaky thus indicating an optimum lower than this. On the other hand 1.32 per cent gave an ash which was not always light enough in color. In the 1931 crop a percentage of 1.55 gave satisfactory color and coherence while 1.96 was too flaky. In the 1932 crop, the only cigars with satisfactory ash characteristics were those with 1.53 per cent magnesia and even these were not as light as might be desired. The results are not altogether consistent, but are sufficient to indicate an optimum between 1.5 and 2.0 per cent of magnesia. It is not likely that any narrower range (more exact location of the optimum point) than this can be determined. In the first place, there is some variation between the percentage in leaves from the same plot. Secondly, the completeness of the com- bustion is probably influenced by the ratio of magnesia to potash quite as much or more than by the actual percentage of MgO in the leaf. The depressing effect of magnesia on the other bases is well illustrated in these analyses. Relation of Calcium to Growth of Tobacco 473 Conclusions and Recommendations These experiments indicate that the optimum percentage of mag- nesia in the leaf is about 1.5 to 2 per cent of the moisture-free weight of the cured leaf. When the percentage falls lower, the combustion is not so complete and a dark ash results. This is accompanied by a less desirable taste and aroma. On the other hand, when the percentage is raised to 2.5 per cent or higher, the ash is whiter but is undesirable because it flakes too much. Of the rates of application tried in this experiment, the 100- pound application came the nearest to keeping the magnesia con- tent of the leaf at the desired percentage during the first season. Since heavier rates of application all gave an undesirable flaky ash in the first year, it seems that the most advantageous practice would be to apply not more than 100 pounds, but to repeat it each year. The practical grower will be guided by the character of the tobacco produced on a field. In many cases, where the ash is already satisfactory, no application of magnesia should be made. If it is too dark, he should apply only the minimum amount that will make the ash satisfactory. In no case should this be more than 100 pounds MgO to the acre. It is not unlikely that a smaller dose applied yearly will usually suffice, especially on heavier soils. The suitable dose may be determined by testing the burn each year on the cigar. The grower must always keep in mind the danger of liming his land so much as to encourage black rootrot. Small applications are not dangerous because the natural leaching of tobacco soils removes the excess. In using magnesian lime, he should purchase a material with the highest percentage of MgO obtainable, keeping in mind that the percentage of calcium oxide is higher in the low-magnesia limes and is almost as effective in raising the soil reaction as is the magnesium oxide. A pure mag- nesium oxide or carbonate might be better, but is not as readily obtained or as cheap as a dolomitic lime. THE RELATION OF CALCIUM TO THE GROWTH OF TOBACCO Calcium, the principal element in liming materials and a con- stituent of numerous materials used as fertilizer, is essential to the growth and development of tobacco and other higher green plants. This element has several functions, or physiological roles, in the plant. Exact knowledge of all these is not yet developed, or is a matter of controversy, but at least three functions are well estab- lished: (i) It neutralizes oxalic and other acids produced within the plant or taken from the soil in surplus quantity. By forming precipitates with the acids it prevents their accumulation, which might otherwise injure the cells. (2) In the form of calcium pectate it is an essential constituent of cell walls. (3) It serves as a vehicle in the translocation of nitrates and probably other anions. 474 Connecticut Experiment Station Bulletin 350 Specific symptoms develop in the green tobacco plant deprived of a supply of calcium sufficient for normal growth. The roots turn brownish with the tips rotted off and the root hairs cease to develop. The entire plant is stunted in growth and in extreme cases the ter- minal buds curl and turn brownish. Tobacco plants grown with- out and with calcium, are shown in Figure 132. Perceutage of Calcium in the Leaf A certain percentage of calcium in the leaf is required for nor- mal growth of tobacco. Garner and co-workers (2) found in Maryland tobacco a minimum calcium requirement of 1.5 to 2.1 per cent CaO in the air-dry leaf. This is in fair agreement with ffoCalcfum MUihCalaum Figure 132. Tobacco plants grown without and with calcium. results of greenhouse tests of Havana Seed tobacco at this Station, which did not show deficiency symptoms when the percentage of CaO was above two. A considerably higher percentage is needed, however, for optimum growth. Although there is no published information with regard to the optimum percentage, an indication is at hand from our own greenhouse tests that about 5 per cent CaO in the air-dry leaf corresponds to optimum growth. Numerous analyses of cured tobacco published in previous reports of this Sta- tion show a wide variation of CaO content ( from about 5 to more than 8 per cent of the dry weight of the leaf). This variation may be governed primarily by the content of calcium (lime) in the soil. Experiments with gypsum at the Windsor station indicate that the CaO content of the leaf increases somewhat in proportion to the lime material added to the soil. Other experiments with pure cal- cium carbonate also show that the percentage increases with increased applications. Relation of Calcium to Growth of Tobacco 475 The Reciprocal Relation of Calcium and Other Bases Many investigators ( i ) ^ have shown that the percentage of cal- cium in the tobacco leaf is decidedly affected by the percentage of potassium and magnesium ; that is, the higher the calcium content the lower will be the content of potassium and magnesium. This was plain in a greenhouse test here in the case of a rather acid soil (4.38 pH) limed with pure calcium carbonate at a rate of i ton and 2 tons to the acre. Pots for quadruplicate treatments were filled with the limed and unlimed soils and were fertilized with ordinary tobacco fertilizer (200 pounds N, 200 pounds K^O, 100 pounds P2O5 to the acre). One plant of Havana Seed was set in each pot and three crops were grown ; for the second crop enough lime was added to bring the reaction up to the same level as for the first crop. The third crop received no additional lime. The tobacco was cured and later analyses were made for lime, magnesia and potash. The results are found in Table 4. Table 4. Percentage of CaO, MgO and K2O in Cured Leaves OF Tobacco Plants Grown With Various Amounts of CaCOs Percentage found in eaf Crop Amount of lime added to soil CaO MgO K2O First u 11 None 1 ton 2 tons 3-31 5-71 6.93 •57 • 33 •25 5-05 4.61 4.18 Second u u None 1 ton equiv. 2 " " 4.64 7.98 8.42 •71 .48 ■38 5^56 4.28 393 Third u None Residual 1.90 2.78 •53 .38 7-49 7-34 3-55 •32 7-13 Here it is seen that in every case where the percentage of calcium was increased, the magnesium and potassium were correspondingly lowered. On the other hand, where the calcium content was con- siderably lower, as in the third crop, the percentage of potassium came up to a higher level. In another section of this bulletin (page 472) where the effect of various quantities of magnesium are reported, it is shown that if the percentage of magnesium was increased in the leaf the calcium content (as well as that of potas- sium) was lowered somewhat in the same proportion. Other experiments here and at New Haven have indicated that sodium (the base in nitrate of soda) and ammonium (the base in sulfate of ammonia) to a certain extent, have a depressing effect on the absorption of calcium. This is in agreement with observa- tions by other investigators (3, 4). ^Figures in parentheses refer to "Literature Cited," p. 478. 476 Connecticut Experiment Station Bulletin 350 The Effect of Various Compounds of Calcium In a greenhouse experiment at the Windsor station where the effect of different calcium compounds on tobacco was compared in both sand and water cultures, it was shown that the compounds varied considerably in availability. Table 5 gives the percentages of CaO found in the leaves when equivalent quantities of CaO were supplied from various com- pounds. In examining these results, we find that calcium nitrate produced the highest percentage of CaO in the leaf and somewhat similar results were found with nitrate and sulfate (gypsum) in combination, while the latter alone produced a considerably lower percentage. Oxide and carbonate, were still less available than the sulfate. Mono- and tri-phosphate, contained in acid phosphate and bone materials, respectively, gave much less CaO in the leaf. Acetate, oxalate, tartrate and citrate, all of which may be present in organic materials, showed low availability of calcium. Table 5. Percentage of CaO in Tobacco Plants Grown in Sand With Various Calcium Compounds Percentage Form of compound CaO Nitrate 5.07 Nitrate — sulfate 4-96 Sulphate 3.55 Oxide 3.53 Acetate 3 • 30 Carbonate 3- 19 Oxalate 2.68 Mono-phosphate 2 . 46 Tartrate 2.19 None 2 . 00 Tri-phosphate 1.92 Citrate .56 Effect on Yield It is commonly known that liming an acid soil will increase the yield of many different crops. The increase in growth usually has been attributed to the neutralizing effect of lime. This may be true in instances where the reaction is very low. Thus, Morgan and co-workers (5) report that very acid soils — below 4.8 pH — have produced increased growth of tobacco through addition of lime. That calcium (the active base in liming materials) as a nutrient is responsible for increase in growth has been shown in field experiments as well as in greenhouse tests at Windsor. Fur- Relation of Calcium to Growth of Tobacco 'ill thermore it has been indicated through other experiments that the activity of calcium is governed by the amount of magnesia present, and vice versa. In other words, there seems to be a Hme-magnesia ratio at which the land will produce optimum growth. Garner (2) states that it takes at least four times as much lime as magnesia to produce the best growth of tobacco on Maryland soil. Similar results have been obtained in our own tests with greenhouse cul- tures. An indication that this ratio governs the growth somewhat is given in the article on magnesia in this bulletin (see page 468). It is shown here that the yield figures are kept at about the same level where varying quantities of magnesian lime were applied, while the ratio of calcium to magnesium was the same. Recent studies here have pointed toward the possibility that an insufficient absorption of calcium by the tobacco plant may be the ultimate cause of brown rootrot on tobacco. This insufficient absorption may not always be brought about by a too low supply in the soil, but by an improper balance between calcium and other bases or a low supply of nitrates, which, as previously stated, would reduce the absorption of calcium. In some preliminary tests, a mixture of equal parts of nitrate of lime and land plaster applied at a rate of 300 pounds to the acre, restored the normal growth to tobacco. Nitrogen is generally considered to be the first limiting growth factor for tobacco in Connecticut. Since nitrogen and calcium are interrelated in their action, as has been pointed out by Parker and Truog (6), both would thus be most important limiting factors in growth of plants. It is, therefore, of importance to consider the calcium need as well as the need of other nutrients in the fertiliza- tion of Connecticut tobacco soils. In general these soils have a low calcium content and furthermore a great deal leaches out of the soil annually. The extent of leaching is reported elsewhere in this bulletin (page 486). It is difficult, however, to lime a tobacco soil to the extent of obtaining sufficient quantities of available calcium, without increas- ing the reaction to a critical point where black rootrot may occur. It is true that some lime is added where magnesian lime is applied to satisfy the need of magnesia (see page 467 of this bulletin) and in cases where a special phosphorus carrier (bone phosphates) is believed necessary. Some calcium is also added through the use of cottonhull ashes. If it is, however, necessary to exclude all these materials from the fertilizer formula, it is a good plan to include some neutral calcium salts such as nitrate of lime or gypsum (land plaster) . The latter material may be applied at a rate of 300 to 500 pounds to the acre and the rate for nitrate of lime should be governed by the amount of nitrate desirable for any particular formula. 478 Connecticut Experiment Station Bulletin 350 Literature Cited 1. Anderson, P. J., Swanback, T. R. and Street, O. E. Potash require- ments of the tobacco crop. Conn. Agr. Expt. Sta., Bui. 334: 143- 146. 1932. 2. Gabner, W. W., McMurtrey, J. E. Jr., Bowling, J. D. Jr. and Moss, E. F. Magnesium and calcium requirements of the tobacco crap. Jour. Agr. Research, 40: 145-168. 1930. 3. HoAGLAND, D. R. and Martin, J. C. Effects of salts on the intake of inorganic elements and the buffer system of the plant. Univ. Calif. Agr. Expt. Sta. Tech. Paper 8. 1923. 4. HoLLEY, K. T., Pickett, T. A. and Dulin, T. G. A study of ammonia and nitrate nitrogen for cotton. Ga. Expt. Sta. Bui. 169. 1931. 5. Morgan, M. F., Anderson, P. J. and Dorsey, Henry. Soil reaction and liming as factors in tobacco production in Connecticut. Conn. Agr. Expt. Sta. Bui. 306. 1929. 6. Parker, F. W. and Truog, E. The relation between the calcium and the nitrogen contents of plants and the function of calcium. Soil Sci., ID : 49-56. 1920. FURTHER EXPERIMENTS WITH NITROPHOSKA* Nitrophoska (No. 3) is a concentrated commercial mixture con- taining 16.3 per cent nitrogen, 16.3 per cent phosphoric acid, and 20 per cent potash. It is a purely chemical mixture without organic material. Because it contains the three important fertilizer elements in a very concentrated form, its use would mean consider- able economy of time and labor if it were found to be suitable for tobacco. Long experience has lead to the commonly accepted belief in Connecticut that a good tobacco fertilizer must contain consider- able organic material and in this respect Nitrophoska is lacking. However, in order to test the soundness of this common belief and at the same time to test the value of this new material, a field experi- ment was started in 1929. On adjacent plots, Nitrophoska alone was compared with a standard formula in which 80 per cent of the nitrogen was from organic material. As a compromise between the two, other plots were treated with a mixture in which one-half of the nitrogen was in Nitrophoska and one-half from the organic materials. Two plots received each treatment and the location and treatment of plots has remained the saine during four years. (See the above mentioned previous reports for composition and quantity of fertilizer.) No diflferences in growth could be observed in the field. The yield and sorting records for 1932 are presented in Table 6. The averages for three years of the treatment are given in Table 7. From the results of this year as well as the average results of three years it appears that Nitrophoska under the conditions of the experiment, consistently produced tobacco of lower grading than the standard formula. There is also a tendency, although less noticeable, toward lower yields through the use of this material. 1 For previous reports on Nitrophoska see Conn. Sta. Buls. 326: 377-379 and 335: 252. Broadleaf Fertilizer Experiments 479 Table 6. Yield and Sorting Records of Nitrophoska Plots. Crop of 1932 Proportion Plot no. Acre yield Percentage of grades Grade index Nitrophoska Plot Ave. L M LS ss LD DS F B Plot Ave. None Half Nitrophoska All Nitrophoska N28 N28-I N29 N29-I N30 N30-I 2070 1974 2016 1866 1957 1839 2022 I94I 1898 7 II II 4 9 5 9 8 ID 5 II 7 35 33 30 32 29 27 0 I I I I I 28 28 27 31 27 32 0 0 0 I 0 0 19 17 19 26 20 25 2 2 2 0 3 3 ■439 .482 •455 .386 •437 .381 .462 .421 .409 Table 7. Nitrophoska Series. Summary of Three Years Results Proportion of Plot no. Acre yield by years Grade index Nitrophoska 1930 193 1 1932 Ave. 1930 1031 1932 Ave. None Half Nitrophoska All Nitrophoska N28 N28-I N29 N29-I N30 N30-I 1884 1829 1813 1856 I915 1875 1793 1764 1813 1856 1813 1820 2070 1974 2016 1866 1957 1839 1886 1870 1870 .491 .464 •457 •453 •435 •473 •493 .481 •451 •478 .440 .446 •439 .482 •455 .386 ■437 • 381 .475 ■447 •435 BROADLEAF FERTILIZER EXPERIMENTS J. S. OWENS^ During the last 10 years the fertilizer experiments conducted by the Tobacco Substation have been concerned with Havana Seed tobacco on soils west of the Connecticut River best adapted to that type. Are the conclusions drawn from these experiments equally applicable to Broadleaf tobacco grown on the fields east of the river on soils that are believed to be best adapted to Broadleaf ? Since the type of soil, Merrimac, is the same on the Station farm as that in the larger part of the Broadleaf region, it has been assumed that the fertilizer response for the two sections would be similar. How- ever, Broadleaf growers at times have been inclined to question this assumption and have made many requests that fertilizer tests similar to those at Windsor be conducted with Broadleaf and on typical Broadleaf soil. It was in response to these requests that the experiments here briefly summarized, were begun. These experiments will be con- * Extension Agronomist. 480 Connecticut Experiment Station Bulletin 350 tinued and extended as far and as fast as funds at hand will allow. The present preliminary report deals with only two questions : 1. How much phosphorus, if any, should be used? 2. What sources of nitrogen are best and most economical? A series of fertilizer treatments was planned in the spring of 1930, and located on the farm of Harry Farnham, East Windsor Hill. The crop was destroyed by hail. The growth had been erratic, apparently because of rootrot infestation and variations in soil. The next season, a location of more uniform soil, better adapted to experiments was found on the farm of Frank Roberts, East Hartford. Plots were laid out and the treatments used in 1930 were continued. The plots were one-twentieth acre in size and wide enough for three rows ; only the center was saved for measurements. Each treatment was applied to one plot in each of three series. Records of each treatment were therefore obtained each season from three plots. Sources of Nitrogen Three sources of nitrogen, cottonseed meal, castor pomace, and urea, were compared on plots where each in turn furnished a large proportion of the nitrogen. The formula with cottonseed meal as the chief source of nitrogen was as follows : Nutrients per acre Materials per acre N PaOs K2O MgO Cottonseed meal, 2425 lbs 160 72 48 24 Nitrate of potash, 295 lbs 40 130 Sulfate of potash, 45 lbs 22 Precipitated bone, 72 lbs 28 Magnesian limestone, 70 lbs 14 Total 2907 lbs. 200 100 200 38 In a second series, castor pomace replaced the cottonseed meal in supplying i6o pounds of nitrogen, and, in the third, cottonseed meal and urea each supplied lOO pounds. The yields and grade indices are given in Table 8. Table 8. Nitrogen Series. Summary of Two Years Results Nitrogen Source Acre yields Grade index 193 1 1932 Ave. 193 1 1932 I519 187I 1695 •495 •447 1632 1948 1790 •444 .469 1599 1859 1729 •434 .469 Ave. Cottonseed meal % Castor pomace %. . Urea J^ •471 •457 •452 Broadleaf Fertilizer Experiments 481 Thus the highest average yield for the two years was on the castor pomace plots, while the cottonseed meal tobacco had a slightly better grading. However, these tests have not been continued long enough to warrant final conclusions. Quantity of Phosphorus Four quantities of phosphoric acid were used, the lowest being only that contained in the 1,740 pounds of cottonseed meal (50 pounds PgOg per acre). Precipitated bone was added to the other plots to make the total amounts of phosphoric acid. The basic formula used was as follows : Nutrients per acre Materials per acre N P2O5 K2O MgO Cottonseed meal, 1740 lbs 114 50 35 17 Urea, 100 lbs 46 Nitrate of potash, 1295 lbs 40 130 Sulfate of potash, 72 lbs 35 Magnesian limestone, 100 lbs 20 Total 2307 lbs. 200 SO 200 2)7 The yields and grading are given in Table 9. The variations, if any, that can be attributed to the phosphorus treatments are slight. The yield on one of the plots with the small- est phosphorus application treatment was so low that the average for the treatment was appreciably changed. However, the lower grade index of the highest phosphorus treatment in 1932 was due mainly to shed damage to the tobacco from one plot. Table 9. Phosphorus Series. Summary of Two Years Results Pounds P2O6 per ' Acre yields Grade index acre 193 1 1932 Ave. 193 1 1932 Ave. 50 ICO 200 300 1593 1589 1661 1631 1876 1936 1902 1923 1735 1763 1782 1777 .440 ■454 •456 .442 ■458 •452 •451 .441 •449 •453 •454 .442 The growth of the 1932 crop was large and uniform throughout the plots. In 1^31, drought affected a portion of one side of the field sufficiently to conceal small differences that might be due to the fertilizers. The variations in both the nitrogen and phosphorus treatments seem to have made but small differences in either yield or quality of the tobacco, as shown by the sorting records. H any conclusion can be drawn at this time it is that all of the treatments grew good crops of Broadleaf. It will be necessary to continue the experi- ment for several more seasons to be certain what effects are impor- tant. 482 Connecticut Experiment Station Bulletin 350 CONSERVATION OF PLANT NUTRIENTS BY COVER CROPS M. F. Morgan^ and O. E. Street In a five-year field experiment at Windsor- on cover crops it was found that the sowing of cover crops each year increased the yield and improved the grading of tobacco. Of the several ways in which such improvement might be efifected through cover crops, it seems likely that the most important is through the conservation of plant nutrients. In order to find out just how much of each nutrient material is actually saved by cover crops, the experiments here described were begun in 1931. When tobacco is grown continuously, without intervening cover crops, the soil surface lies bare for nine and a half months of the year. Severe leaching usually occurs during fall and spring months, and during mild winter periods. Since not more than two-thirds of the nitrogen applied in fertilizer of the usual tobacco formula, is accounted for in the crop removed, there is a residue of fertilizer nitrogen, in addition to the more slowly available nitrogen reserve of the soil at the time of tobacco harvest. Much of this is transformed to nitrates during the comparatively warm period from August 15 to November i. When no cover crop is seeded after the tobacco is removed, all of the valuable product of this nitrification process is leached from the soil before the following season. The nitrates take with them other valuable nutrients, cal- cium, magnesium, and potassium, in their downward escape from the soil. If a cover crop is seeded after tobacco, much of the nitrogen becoming available as nitrates during the fall period is taken up by the crop. The more luxuriant the fall growth, the more efficiently are nitrates withdrawn from danger of leaching. Also, the grow- ing cover crop removes much moisture, hence a smaller portion of the rainfall washes down through the soil. If, as in the case of oats, the cover crop dies at the onset of win- ter, the dead residue of stems and leaves lies upon the soil surface until plowed under in the spring. During this period most of the potassium and a good part of phosphorus in the above ground resi- due may be leached into the soil, as has been shown in studies of changes in the chemical composition of dead leaves exposed to the weather during the fall and early winter period. However, these elements are not readily leached from the soil. The nitrogen, cal- cium and other mineral elements in the dead crop residue are not ' liberated to any appreciable extent until decomposition begins in the spring, particularly after the crop is plowed under. If the cover crop is winter-hardy, the nutrients taken up by the crop are retained in the living parts until plowed under. * Agronomist in charge of Soils Department. ^ Final report in Conn. Agr. Expt. Sta. Bui. 335: 227-231. Conservation of Plant Nutrients 483 These statements could be made as a result of logical deductions from established principles, but a quantitative measure of the value of cover crops in conserving nutrients against leaching losses is now supplied by a series of lysimeter trials established at Windsor in the spring of 1931. Data are now available for the year 1931-32 and the period from May 26 to November 25, 1932. Drainage water from the following treatments has been collected and analyzed. Tank Nos. Fertilizer Crop 2II-2I2 No nitrogen Fallow 213-214 " " Tobacco, no cover crop 215-216 Nitrogen as calurea Fallow 217-218 " " " Tobacco, no cover orop 219-220 " " " Tobacco, oats cover crop 221-222 Windsor N-i tobacco (1/5 of nitrogen as soda, 4/5 of nitrogen ics) formula nitrate of as organ- Tobacco, no cover crop 223-224 " " " " Tobacco, oats cover crop 225-226 " " " " Tobacco, rye cover crop 227-228 " " " " Tobacco, timothy cover crop Nitrogen is applied at the rate of 200 pounds per acre per year, phosphoric acid at the rate of 100 pounds, potash at the rate of 200 pounds, and magnesia at the rate of 50 pounds are applied to each tank. The tanks are 30 inches deep and contain 8 inches of surface soil and 2cr inches of subsoil. Quantity of Water Leached The growth of cover crops has exerted a significant effect upon the quantity of water leached during the fall and early winter period. This is shown in Table 10. The greater effect of the cover crops sown in 1932 is probably due to the more luxuriant growth that resulted from more favorable moisture conditions. The 1931 season was unusually dry during late summer and fall. On the basis of these results, it is evident that oats withdraw water from the soil to a greater extent than either rye or timothy during its short period of active growth. Temporary Withdrawal of Nitrogen by the Cover Crop In an effort to reveal the amount of nitrogen taken up by the cover crop, the green oats plants, including all of the fine roots which could be separated from the soil, were collected from four typical areas in the fields at Windsor, each 4 by 4 feet in size. 484 Connecticut Experiment Station Bulletin 350 Plants and heavy roots of the oats cover crop weighed (oven dry) 1204 pounds per acre on October 21, 1931, and contained 29.31 pounds of nitrogen. Fine fibrous roots weighed 796 pounds per acre, and contained 18.4 pounds of nitrogen. The nitrogen thus accounted for in the Hving oats plants on that date amounted to 47.71 pounds per acre. The crop grew slowly until the middle of November before being killed by cold weather. It must be kept in mind that the cover crop of 1931 was lighter than usual, due to dry weather conditions. Table 10. Effect of Cover Crop Upon Quantity of Water Leached During Fall and Early Winter Period (in Acre Inches). Treatment Fallow, no nitrogen Tobacco, no nitrogen, no cover crop Fallow, nitrogen as calurea Tobacco, nitrogen as calurea: No cover crop Oats cover crop Tobacco, Windsor N-i formula: No cover crop Oats cover crop Rye cover crop Timothy Period Sept. I, '31-Feb. I, '32 Total leaching 5.686 4-560 5-803 3-994 2.512 4.282 2.675 3-007 3-840 Decrease for cover crop 1.482 1.607 1-275 0.442 Period Sept. I, '32-Jan. I, '33 Total leaching 8.015 7.900 7.845 7-445 4.618 7-795 4.623 6.212 6.861 Decrease for cover crop 2.827 3.172 1.583 0.943 The ainount of nitrogen taken up by the oats cover crop in 1931 was also determined by comparing the nitrate nitrogen content of the soil on October 21. There had been no leaching of nitrates from the surface soil since the sowing of the cover crop. The average of four samples under the oats crop showed approximately 16 pounds of nitrate nitrogen per acre in the surface soil. Four samples from an adjacent plot without cover crop gave 81 pounds on the same basis. The oats crop had caused a decrease of 65 pounds of nitrate nitrogen per acre in the surface soil. The amount recovered in the oats was somewhat lower than this last figure, due to the failure to separate a part of the fine fibrous roots from the soil. Conservation of Plant Nutrients 485 Effect of Cover Crop on the Leaching- of Nutrients The results of drainage water analyses for important nutrient elements are shown in Tables ii, 12, 13 and 14. The data for the six-months period May 26, 1932, to November 25, 1932, does not give a complete picture for the year. However, because of very heavy autumn precipitation, the leaching of practi- cally all of the nitrates present in the soil during that period, was assured. Concentration of the leachates collected on January 3, 1933, did not exceed 10 parts per million of nitrate nitrogen on any of the cropped tanks, and had fallen to less than one part per mil- lion under oats and rye cover crops. The value of the oats cover crop in conservation of nutrients is demonstrated by averaging the data for the two types of fertiliza- tion for two seasons, as shown in Table 15. Rye conserved the nutrients to about the same degree, while timo- thy was significantly less effective in preventing losses through leaching. There were no consistent differences in the amounts of other elements leached from the soil as a result of cover cropping. Although while it is difficult to evaluate fairly the monetary value of residual plant nutrients, it is worth bearing in mind that the above figures represent the equivalent of at least $7.(X) worth of nitrate of soda, 75 cents worth of sulfate of potash, and 50 cents worth of dolomitic lime, at 1933 fertilizer prices. One must also take into consideration that a retention of this amount of nitrogen is associated with a conservation of 1000 pounds per acre of organic matter, or as much as would be supplied in 2 tons of manure. Table ii. Leaching of Nitrogen as Affected by Cover Crops (Pounds per Acre) Treatment Fallow, no nitrogen Tobacco, no nitrogen, no cover crop Fallow, nitrogen as calurea Tobacco, nitrogen as calurea: No cover crop Oats cover crop Tobacco, Windsor N-i formula; No cover crop Oats cover crop Rye cover crop Timothy cover crop May 26,'3i-May2S,'32 Leached 157-69 64-51 316.14 71.69 24.90 81.14 21.69 18.89 28.19 Retained by cover crop 46.79 59-45 62.25 52.95 May 26,'32-Nov. 25/32 Leached 72.75 33-53 218.01 68.61 2.67 55-88 4-94 1-38 27.61 Retained by cover crop 65-94 50.74 54-50 28.27 486 Connecticut Experiment Station Bulletin 350. Table 12. Leaching of Calcium as Affected by Cover Crops (Pounds Per Acre) Treatment May 26,'31-May 25, '32 Leached Retained by cover crop May 26,'32-Nov. 25, '32 Leached Retained by cover crop Fallow, no nitrogen Tobacco, no nitrogen, no cover crop Fallow, nitrogen as calurea Tobacco, nitrogen as calurea: No cover crop Oats cover crop Tobacco, Windsor N-i formula: No cover crop Oats cover crop Rye cover crop Timothy cover crop 186.44 106.41 351-57 120.61 72.55 101.41 62.16 64-95 74.62 48.06 39-25 36.46 26.79 87.41 49-45 206.98 76.89 19.62 47-36 15-60 16.10 27-75 57-27 31-76 31.26 19.61 Table 13. Leaching of Potassium as Affected by Cover Crops (Pounds per Acre) Treatment Fallow, no nitrogen Tobacco, no nitrogen, no cover crop Fallow, nitrogen as calurea Tobacco, nitrogen as calurea: No cover crop Oats cover crop Tobacco, Windsor N-i formula: No cover crop Oats cover crop Rye cover crop Timothy cover crop May 26,'31-May 25, '32 Leached 132.53 89.85 198.84 72.10 61.55 88.53 67.16 66.00 72.77 Retained by cover crop 10.55 21.31 22.53 15-77 May 26,'32-Nov. 25, '32 Leached 84.60 49.66 126.97 54-39 23-52 56.46 24.94 28.80 43-88 Retained by cover crop 30.87 31.52 27.66 12.58 Conservation of Plant Nutrients 487 Table 14. Leaching of Magnesium as Affected by Cover Crops (Pounds per Acre) Treatment Fallow, no nitrogen Tobacco, no nitrogen, no cover crop Fallow, nitrogen as calurea Tobacco, Nitrogen as calurea: No cover crop Oats cover crop Tobacco, Windsor N-i formula: No cover crop Oats cover crop Rye cover crop Timothy cover crop May 26,'31-May 2S.'32 Leached 29.00 22.57 61.40 23.67 10.70 15.12 10.71 11-73 18.85 Retained by cover crcp 12.97 4.41 3-39 May 26,32-Nov. 25, '32 Leached 14.29 6.83 41.15 12.45 4.21 8.76 1.99 I.9I 2.74 Retained by cover crop 8.24 6.77 6.85 6.02 Table 15. Average Conservation of Nutrients by the Oats Cover Crop Nutrient Nitrogen. . . Calcium. . . Potassium. . Magnesium . Pounds per acre 55-73 44.08 23-56 8.09 Equivalent Oxides 61.68 28.38 13-41 Lysimeter Results at Other Stations In lysimeter studies made at other stations, no strictly com- parable tests have been recorded, but the following data from the Cornell lysimeters strengthen the conclusions drawn from our own findings. Table 16. Nutrients Leached from Cornell Lysimeters (Nos. i to 12, Period of 1910-1919) Annual drainage loss per acre (pounds) Treatment Nitrogen Calcium Magnesium Potassium Fallow Permanent sod 69.0 2.5 6.7 398 260 246 63 50 43 72 62 Rotation without legumes 61 488 Connecticut Experiment Station Bulletin 350 At the New York State Agricultural Experiment Station, Geneva, an early series of lysimeters (1884- 1886) showed an annual drainage loss of 195 pounds of nitrogen per acre from fal- low soil, while permanent sod permitted the leaching of only 0.67 pounds per acre per year. In more recent experiments at Geneva, in a comparison between a rotation of alfalfa-barley-wheat and two years of alfalfa followed by two years of fallow, the soil under the grain crops leached 123 pounds less nitrogen per acre as an annual average of the six years. In this case, however, the crops were harvested. TOBACCO INSECTS IN 1932 Donald S. Lacroix Prevalence of Various Species The eastern field wireworm, Pheletes eciypus Say, caused quite as much damage to newly transplanted tobacco as in 1930, and the effects of the activities of this insect were apparent until July. This is unusual, as wireworm injury normally occurs only over a period of two to three weeks, from late May until about the middle of June. Adults of this insect were flying in small ntunbers during the last two days in May and the first week in June. Many fields that suffered last year had a light infestation this year, or none at all. The potato flea beetle, Epitrix cucumeris Harr., was much less in evidence during the 1932 season than in 193 1. As usual, the insect was present in practically all fields visited, but injury caused by it was generally light. The tobacco flea beetle, Epitrix parvula Fabr., was found in small nimibers on shade grown tobacco at the Station in Windsor. This is the first record of the occurrence of this species in Connecti- cut. Tobacco horn worms, Phlegethonfius quinquemaculata Haw., and P. sexta Johan., appeared in their usual numbers. Broadleaf tobacco in the eastern part of the tobacco-growing district was injured most. The tobacco thrips, Frankliniella fusca Hinds, occurred through- out the Connecticut tobacco areas on Broadleaf, Havana Seed and shade tobacco, causing considerable injury to lower leaves (Figure 133). This was in direct contrast to last year's condition, when only a few fields bore evidence of the insect. The stalk borer, Papaipema nitela Guen., was reported from only two plantations. The tarnished plant bug, Lygus pratensis Linn., appeared in its usual niimbers and caused considerable damage. More about this insect will be found on page 493. Tobacco Insects in iq^2 489 The Mexican bean beetle, Epilachna corrupta Muls., was taken on tobacco leaves during late July and eariy August. To see whether the insect actually would feed on tobacco, several larvae were placed on a plant. They promptly migrated without feeding. The tobacco budworm, Helioihis virescens Fabr., appeared on both shade grown and sun grown tobacco this season in Avon, East Hartford, Poquonock, and Windsor, but in such small num- bers that injury was not extensive. More concerning the Hfe history Figure 133. Thrips injury (right) to leaf of Havana Seed tobacco as compared with normal leaf (left). and habits of this southern species appears on page 491 of this pub- lication. Grasshoppers of various species were about as numerous as in 1930 and 1931 on sun grown tobacco. Chief among these was the red legged grasshopper, Melanoplus femur-rubrum De G. Potato Flea Beetle^ Studies of the habits and activities of the potato flea beetle on tobacco were continued this season. Eggs were deposited in rearing jars at different times from late June through early August, and ' Epilrix cucumeris Harr. 490 Connecticut Experiment Station Bulletin 350 hatched in five to eight days. Due to the fact that difficiilty was experienced in getting mature larvae from eggs laid in rearing, no reliable figures as to length of larval life can be given. Larvae taken from potato fields pupated in rearing jars and emerged in seven days. No larvae or pupae were found in potato fields after July 29. The first adults of the season were taken in the seed beds in the middle of May. No larvae or pupae were found this season in tobacco soils, either in seed beds or out in the field, although there was an abundance of larvae and pupae in potato soil about 400 yards from the tobacco. Repeated siftings were made throughout the summer in order to obtain larvae or pupae. Although this may seem to indicate that the adults migrate from the outside onto the Table 17. Flea Beetle Population on Station Tobacco, 1932. Number of flea beetles on 25 plants Date Sun grown Shade grown Total Section 1 Section 2 Section 3 Section 5 June 18 52 48 6 2 108 « 25 12 40 9 0 61 " 28 16 66 5 2 89 " 30 25 74 6 I 106 July 2 18 74 0 0 92 6 13 26 4 I 44 8 19 32 6 7 64 " 12 26 39 II 8 84 « '4 19 42 8 9 78 " 16 32 33 7 II 83 " 20 39 31 6 3 79 " 23 28 33 61 " 27 13 21 34 tobacco, it must be borne in mind that the infestation on Station tobacco during 1932 was unusually light. The population of adult beetles on the Station tobacco is pre- sented in Table 17. A comparison of this table with those included in the reports for 1930 and 1931 will show a marked decrease in beetle abundance for the past season. Control. After trying barium fluosilicate for three seasons against the flea beetle, it is concluded that this material, used as a dust, is a very satisfactory way to combat this pest. The use of it during 1930 and 1931 showed conclusively that it was of value in holding the flea beetle in check, but an undesirable residue was left. To obviate this, it was necessary to reduce the amount and use it undiluted. Applying barium fluosilicate at a rate of 4 to 5 pounds to the acre resulted in good control, and four dustings at Tobacco Insects in IQS- ^^1 weekly intervals left no visible residue. Three different plots were treated two on sun grown tobacco and one on shade grown. One of the sun grown plots was treated three times, once every two weeks, and showed as good results as did the other two. Several growers in Connecticut used this material during the season of 1932 with satisfactory results. One used it at a rate of 3 pounds to the acre, diluting with tobacco dust to make it more bulky, and got very good control with a single application. In seasons of light infestation, relatively few applications of dust are necessary. In seasons of heavy infestations, four or five weekly applications during the month of July should be sufficient to hold the beetles in check. It is essential that the dusting be done during the part of the day that is most calm, either early in the morning or in the evening. Due to the fact that the tobacco leaf possesses glandular hairs, which secrete a gummy substance, the dust will stick to the foliage very well, and therefore it is not neces- sary to dust while dew is on the leaves. Number of Flea Beetles per 25 Plants on Treated and Untreated Plots Treated Not Treated July 8 July 21 July 8 July 21 Plot I, Sun grown 4 2 Check 19 31 " 2, Shade grown o i Check 7 3 " 3, Sun grown i 3 Check 32 39 An examination of the plants on July 26 indicated no injury from beetles or dust on the treated areas, but some beetle injury on the untreated areas. Tobacco Budworm^ This insect, which causes injury of great importance to tobacco grown in the south, appeared in several fields in Connecticut this summer. Although quite widely distributed here, it did compara- tively little damage. The first infestation that came to our atten- tion was in Avon early in July. Type of injury. The worms, when immature, feed on the buds (growing tip) and suckers, eating holes in the leaves as they unfold. As the leaves grow larger, the holes also become larger, and a mal- formed leaf results. The more mature worms feed on larger leaves. Life history and appearance. On July 5, 1932, the larvae in the field were in all stages from newly hatched individuals to half- grown worms. The smallest ones were about one-eighth inch in length and rusty brown in color. The half-grown ones were about three-fourths inch long, pale green, with several black tubercles on ' Heliolhis virescens Fabr. 492 Connecticut Experiment Station Bulletin 350 each segment, from each of which sprouted a single hair. Usually only one larva could be found on each plant, rarely two. These larvae were placed in rearing jars and reached maturity about the middle of July. At the mature stage the larvae were approximately i ]/^ inches long, green, with stripes running length- wise along the sides, and with sparse hairs (Figures 134 and 136). On July 17, 18, and 19 the mature larvae left the foliage and went below the soil surface, where they rested for four days. At the end of this "rest period" pupation took place (July 23, 24, and 25). The pupa, Figure 135, is at first greenish, and spindle-shaped, but it Figure 134. foliage. Tobacco budworm larvae and their injury to tobacco turns gradually to a mahogany brown. About 10 days are spent in the pupal stage. From this emerges a small moth with light green wings. Each front wing is crossed by four paler oblique bands (Figure 135). On August 25, half to two-thirds of the grown larvae were found boring into seed pods of some shade tobacco that was not grown under cloth (Figure 136). Several of these pupated during the first week in September and adults emerged from late in September through the middle of October. This insect has been previously reported in Connecticut (Report of Connecticut State Entomologist for 1909-10, pp. 367-368) as Tobacco Insects in /pj^ 493 feeding on tobacco September 15, with moth emergence October 15-21. It seems possible then, for two generations to occur here. Control. In the south, the budworm is held in check by apply- ing one pinch of a mixture of i pound of lead arsenate to 7 5 pounds of com meal on the bud of each plant. This is usually done by hand. Unless the insect becomes more abundant than it did in 1932, hand picking is the most satisfactory method of control here. Red Arrow spray at a dilution of i to 500 was tried as a control measure for bud worms, but failed to kill them at this strength Horn worms and climbing cutworms treated with the same material and at the same time, died shortly Figure 135. Pupa (left) and adult moth (right) of tobacco budworm (twice natural size). In one field of Havana Seed tobacco, signs of early feeding were found, but no signs of later damage were visible. A careful exam- ination showed that parasites had been at work and had killed the budworm larvae. The parasite cocoons were found fastened to the leaf with parts of the dead worm still hanging to the cocoons. The adults that emerged from the cocoons have been identified as Sagaritis provancheri Dalla Torre (dubitatus Cresson) of the family Ichneumonidae. • Tarnished Plant Bug^ Ts^pe of injury. The tarnished plant bug pierces the growing tip of the tobacco plants and sucks juices from the bud tissue. As a result of this feeding, the unfolding leaves become twisted and curly in appearance. Malformed plants were found on many plantations growing Havana Seed. Appearance and habits. The adults are about one-fourth ' Lygus pratensis Linn. 494 Connecticut Experiment Station Bulletin 350 of an inch long, flat, and oval in outline with a small head. They vary in color, usually being brown with spots of black and reddish- brown. They spend the winter under trash, stones, bark of trees, and similar concealed places and appear early in the spring, par- ticularly on weeds, flowers, and grass. Flying is rapid when dis- turbed. Observations made in Windsor during the past two seasons in- dicate that they appear on tobacco only to feed, as no eggs or young bugs could be found. Several times during the season of Figure 136. size). Tobacco budworm boring into seed pod (twice natural 1932 adults were caged on single plants. They all died in time, without depositing any eggs, but seemed to feed freely. Control measures. It is almost impossible to reach the adults with a contact spray, since they are very active and fly away rapidly as soon as one approaches within four or five feet of them. A stomach poison is of nO value in controlling this type of insect. Keeping down weeds and cleaning up all trash near the fields may be of some help in reducing their numbers. Distribution of Wireworm^ Larvae in Tobacco Soil Very little information is available concerning the habits and activities of wireworms in tobacco soil. To obtain some of this very necessary information, careful observations were made on the ' At least two species were present, Pheleles ectypus Say and Limonius plebejus Say. Tobacco Insects in 19^2 495 distribution of larvae in the soil of an infested field in East Granby throughout the season of 1932. At intervals during the summer and fall, soil areas 2 feet square and 2 feet deep were excavated at two different parts of the field. Each area was located so that one-half of the soil was removed from the tobacco row directly beneath the roots, the balance be- tween the rows and away from the roots. Soil to the depth of 3 inches was removed and examined and its larval content noted. Successive soil horizons were treated in a similar manner. Table 18 indicates the results of these periodic soil examinations. An analysis of these data indicates that. throughout the growing season, the largest number of larvae were in the tobacco rows, and also that during the summer months the greatest percentage of larvae, as far as depth is concerned, was from 6 to 9 inches below the soil surface. During these observations it was found that wireworms feed all summer on the roots and bases of stalks, but the plants were grow- ing so fast that this injury was not always apparent above ground. Heretofore it has been assumed that feeding ceased in June, and that the larvae went down to escape heat. An examination of the data presented in Table 18 will disclose that the greater percentage of larvae are at a depth of from 6 to 9 inches dimng the heat of summer, but in the cooler months of October and November, the population seems to move upward, and at this time of year the greatest number of larvae occur in the 3 to 6 inch layer. Just why this condition exists is not readily explained, as no indications of feeding could be found on cover crops that had been sowed im- mediately after the tobacco was harvested. As soon as the crop was removed, the stalks were cut and har- rowed in. Two days later, many larvae could be found in pieces of stalks on or near the surface in spite of the fact that the greatest percentage of larvae was still at the 6 to 9 inch level below the surface. Pupae of Limonius plebejus Say (Figure 137) were first taken late in July at the 6 to 9 inch level. Adults (Figure 137) of this species were not found in the field, but emerged late in August from pupae removed to the insectary. The fact that about two-thirds of the larval population was found to be concentrated in the tobacco rows immediately below the plants suggested the possibility of obtaining some means of control by applying insecticides in the row. Several tests were conducted, based on this suggestion, but the results were unsatis- factory. Another series of experiments was conducted to see if there was preference on the part of wireworm larvae for any particular cover crop. Plots of wheat, oats, barley, rye, and timothy (the usual cover crops planted in the fall on tobacco soil) were sown on an infested field, but the fall feeding of the worms was negligible so that no accurate figures were obtained. 496 Connecticut Experiment Station Bulletin 350 Table i8. Distribution of Wire worm Larvae in Soil of Tobacco Plantation, East Granby. Season of 1932. Number larvae in soil Date Depth In row Between rows Total Per cent June 29 0"- 3" 3"- 6" 6"- 9" 9"-I2" 12 "-24" 5 3 8 6 0 2 7 4 I 0 7 10 12 7 0 19 28 33 19 0 22 61% H 39% 36 July 8 0"- 3" 3"- 6" 6"- 9" 9"-I2" 12 "-24" 2 3 ID 6 0 2 3 4 0 0 4 6 14 6 0 II 20 46 20 0 21 70% 9 30% 30 July 18 0"- 3" 3"- 6" 6"- 9" 9"-I2" 12 "-24" 2 12 17 I 0 I 6 II 0 0 3 18 28 I 0 6 36 56 2 0 32 64% 18 36% 50 July 28 0"- 3" 3"- 6" 6"- 9" 9"-I2" 12 "-24" 3 3 5 6 0 0 3 6 (i Pupa) I 0 3 6 II 7 0 II 22 40 26 0 17 63% 10 37% 27 Aug. 9 0"- 3" 3"- 6" 6"- 9" 9"-I2" 12 "-24" 0 2 8(3 Pupae) II I 0 2 5 2 0 0 4 13 13 I 0 12 42 42 3 22 71% 9 29% 31 Aug. 18 0"- 3" 3"- 6" 6"- 9" 9"-I2" 1 2 "-24" 0 3 ID (4 Pupae) 7 0 0 I 3 (I Pupa) 4 (i Pupa) 0 0 4 13 II 0 0 14 46 39 0 20 71% 8 29% 28 Sept. 10 0"- 3" 3"- 6" 6"- 9" 9"-I2" 1 2 "-24" 0 0 7 8 0 I 0 4 2 0 I 0 II ID 0 4 0 50 45 0 15 68% 7 31% 22 Tobacco Insects in iq^2 497 Table i8. Distribution of Wireworm Larvae in Soil of Tobacco Plantation, East Granby. Season of 1932. — Continued. Number larvae in soil Date Depth In row Between rows Totals Per cent Oct. I 0"- 3" 0 0 0 0 3"- 6" 6"- 9" 6 3 3 4 9 7 52 41 . 9"-I2" I 0 I 5 12 "-24" 0 0 0 0 ID 58% 7 41% 17 Nov. 5 0"- 3" 0 0 0 0 3"- 6" 6"- 9" 7 5 5 3 12 8 54 38 9 "-12'' 0 I I 4 12 "-24" 0 0 0 0 12 54% 9 42% 21 Insecticide Tests Against Wireworms Naphthalene. This material, in the powdered form and mixed with Kayso and water, remained in suspension well and gave some promise in wireworm control last season. It was given another trial at a dilution of 16 teaspoonfuls to i gallon of water along with 4 teaspoonfuls of Kayso. This mixture repelled the worms, but burned the plants. Faradichlorobenzene. This insecticide has been used suc- cessfully in controlling the peach tree borer, clothes moths, and other insects. Several different tests were conducted with it in 1932 against wireworms, and thus far it has proved more satisfactory than any of the other materials tried. The material was placed in the holes dug to receive transplants, water poiured in, and the plants set. At the rate of i teaspoonful to a plant, severe burning resulted. A half teaspoonful caused much less plant injury, and kept the wireworms away. When the material was placed on top of the soil (i teaspoonful to a plant) even less injury was noted unless it came in direct contact with the stalk, and some control of wire- worms was noted. Broadcast over the soil at a rate of 10 pounds to the acre and harrowed in, this insecticide had Httle or no effect on the insect and did not hurt the plants. Another series of tests was conducted to determine the tolerance of the plant, in which i part of paradichlorobenzene was mixed 498 Connecticut Experiment Station Bulletin 350 with lo parts of sand, and sown in a shallow furrow in the soil. The plants were set in this fiirrow One row was treated at the rate of ^ of a pound of this mixture to 36 plants, a second row of i3^ pounds to 36 plants, and a third row treated with 3 ounces of paradichlorobenzene to 36 plants (drilled in before planting) and two rows were left untreated as checks No ill effect was observed on any of the treatments. Other tests were conducted in which the material (about V2 teaspoonful to a plant) was placed directly in the soil with the transplant in a field heavily infested with wireworms. About 10 days later an investigation showed that there were few or no wire- worms within one foot of plants so treated, and in a few instances dead larvae were found near the roots. Injury to the plants was Figure 137. Pupae (left) and adult beetles (right) of wireworms, Limonius plebejus Say, twice natural size. slight and only occasional. The characteristic odor of the insecti- cide was strongly evident in the soil 1 2 days after its application. Barium fluosilicate. This material mixed with water at a rate of 15 teaspoonfuls to the gallon and poured into the soil before setting had no effect on either the plants or the wireworms. Control of Thrips The tobacco thrips mentioned on page 488 was controlled satis- factorily with Pyrethrol and water (1-200). Nicotine sulfate and water (3 ounces to 5 gallons) with 3 ounces of liquid soap gave about 30 per cent kill, and when used with Penetrol (1-200) gave from 50 to 75 per cent kill. An examination of the leaves showed that thrips kept awayfrom the Pyrethrol and nicotine-penetrol plots for two days, whereas they were observed in small numbers on the nicotine sulfate-soap plot 12 hours after the application. Three Studies of Fuels for Curing 499 sprayings with Pyrethrol (at the rate stated) five days apart should reduce injury from thrips to a minimum. COMPARATIVE STUDIES OF FUELS FOR CURING IN 1932 Natural lump charcoal (made by burning wood with only a scanty supply of air) is commonly used to promote optimum conditions in the shed for curing. In recent years, however, processed charcoal, a by-product of certain manufacturing industries, has come into the market. Although these processed materials cost more per ton than lump charcoal, the dealers claim for them certain advantages which compensate for the increased cost. In order to determine to what extent these claims are warranted and to get actual data on their heating value, comparative tests of lump charcoal and two of these processed materials. Ford Briquets and Eastman Charkets, were conducted in shed compartments at the Tobacco Substation in 1932. These compartments are 16 feet square and the height of the shed. Each compartment was filled equally with shade tobacco, four fires placed in each chamber, and the firing commenced as soon as the tobacco was hung. Accurate records were kept of the weight of fuel used and at the end of the firing period the unburned fuel in the pits was screened and the net consumption computed. The tests were made at three periods in the curing season, with the fuels rotated among the three com- partments, in order to overcome any differences in heat loss due to position of the compartments, area of the outside wall, or tightness of the walls. Using a sling psychrometer, records of temperature and humidity in each chamber and outside of the shed were taken at hourly inter- vals during the entire periods of the test. Observations on the uniformity of combustion, ease of starting the fires, labor of main- taining the fires, smoke production, fire hazard through sparks, and proper manipulation of the fires were made from time to time. In the first two trials laths from each chamber were tagged as random samples. After the tobacco had been sweated these sam- ples were sorted and the grade index computed. Fuel consumption and temperature records for the three trials and a summary of the complete records are given in Table 19. The ajverage temperatures maintained were not as high as would be obtained under ordinary shed conditions because the relative wall space per fire is much greater in small compartments. However, it was felt that the precautions taken to insure uniformity of con- ditions made the tests comparable. The last trial was refiring on tobacco of the same picking as the second trial and lasted for only 24 hours. In averaging results, due weight was given to this fact. In each trial the least efficient chamber was No. i, which was characterized by either lower temperature or greater fuel consump- 500 Connecticut Experiment Station Bulletin 350 J3 m to Kl ii t. t-. 3 3 3 O O O J *" ^ A ^ 00 00 "^ '^ rf N □ • o « o i-c rhCS fO CO ■* O 0< ID O OnvO 00 rC lO ^^d. r(- P) ro tfi fe'B io^ n „' 00 o PS > 3 ?J tv. t^ o o p^ t3 as . ^li^rt- fD^O ■* vD VO t^ H Aver chain tem ^Tt-t-^ o d r^ vo M <0 ft! 00 00 00 OvOnQO 00 00 00 w Ol H '6 Q di lOiO ITS lO iz; "5 G M <; r=^ 5-^ ro a^ •* t^f<^-^ t^ •* (O "^ g- \0 « « ^OT^c CJ CO «i S M C0>0 N O lO lO \0 lOlO ^^ ,^ crt o r) lUrc; R 3 rr U ^ n o a pq E S -t-> tn 3 O ca JfeW Studies of Fuels for Curing 501 tion, or both. Other compartments were about equally economical of fuel and heat. The summary discloses that lump charcoal had the highest consumption, but also maintained the highest average temperature under the conditions of this experiment. Eastman Charkets had the lowest consumption and the second highest gain in temperature, while Ford Briquets, although practically as efficient in fuel consumption as Charkets, was the lowest in average tem- perature maintained. In the ease of management of the fires, both of the processed charcoals were much superior to lump charcoal. Starting the fires was not difficult with any of the fuels. The Charkets smoked the least, when the kerosene was applied, because of the less absorptive nature of the product. The temperature increased most rapidly with lump charcoal, but also dropped most quickly and consequently required much more attention. During the night it was necessary to tend the lump charcoal fires at least every hour and the other fuels every two hours. During the day, the processed fuels need replenishing every six hours as compared with two hours for char- coal. Because of the much higher volume weight of the processed fuels, the bulk of fuel applied had to be very carefully watched or the fires would go too high. The best method, especially during the day, was to apply a counted number of the Charkets or Briquets. It was also necessary, with the processed fuels, to use care that unburned fuel did not remain at the bottom of the pits. This was obviated by using a homemade wire fork and overturning the fires entirely at intervals. Fire hazard was much reduced by use of the processed fuels and this is one of the greatest advantages of these fuels. Because of the uniform and compact nature of these briquets, there are no sudden rpinor explosions with consequent scattering of sparks. The percentage of grades and grade indexes for the tobacco from the first trial are presented in Table 20. The grading of the tobacco from the latter tests is not given, since the tobacco was fourth picking and all treatments showed poor grades with no sig- nificant differences. No significant difference in the grade indexes of the tobacco cured with the two processed fuels was found, but the lump char- coal was distinctly inferior in this respect. The most plausible explanation of this difference is the lack of uniformity of tempera- ture with lump charcoal. The large volume of charcoal that must be applied tended to smother the fires, and a temperature fluctuation of 5 to 10 degress often occurred, especially under unfavorable weather conditions. These fluctuations of temperature were accompanied by much wider fluctuations of relative humidity and consequent changes in the rate of water loss and chemical activity, with the result that the tobacco lacked uniformity of color and yel- low shades predominated. 502 Connecticut Experiment Station Bulletin 350 Table 20. Sorting Record of Shade Tobacco (Trial i) Type of fuel Percentage of grades Grade L LL LC LC2 XL Br index! Lump charcoal Ford briquets Eastman charkets 7 15 13 22 29 35 47 33 26 10 9 14 7 II 9 7 3 3 2.89 3-16 3-15 'See BuL 334, p. 178, for explanation of grade index. The comparative values for the difiFerent grades of shade tobacco were as follows: L 5.00 LC2 1.50 LL 4.00 XL 1. 00 LC 3.00 Br •50 SHADE CUBING EXPERIMENTS IN 1932 The experiments conducted in small curing chambers during the three previous years having indicated rather definitely that a tem- perature of 90° F. with a relative humidity of 70 to 80 per cent maintained constantly would produce good colors on first and second pickings, it was decided to study the efifect of comparable conditions in the shed. Through the courtesy of the Gershel-Kaf- fenburgh Tobacco Company, hygrothermograph records of the curing periods in four sheds were obtained and the effect of the treatments correlated with the quality of the tobacco. Preliminary Runs The preliminary run on first picking was started on July 13 and records obtained until July 27. The first firing period was 48 hours, at an average temperature of 91° F., and average relative humidities of 50 to 60 per cent, 35 to 45 per cent, and 30 per cent for successive 16-hour periods. The leaves were definitely in the yellow green, wilting was thorough, and some tips near the fires were becoming husky by the end of this period. The second firing period, of 14 hours duration, following a moist spell, was character- ized by a rapid drop in shed humidity to 30 per cent, which indi- cated that very little of the original water content of the leaves was present, and that the only effect was the removal of the absorbed water. Further control in this shed consisted only of operation of the side ventilators. The preliminary run on i^^ picking was started July 20, and records obtained for nine days. The initial firing period covered 84 hours at an average temperature of 90.2° F. At no time did the relative humidity drop below 40 per cent. Somewhat lower temperature in the final phase of the firing and high outdoor humid- ity tended to keep the shed humidity at a higher level. A' second Studies of Fuels for Curing 503 firing of 12 hours' duration and a third of 6 hours were employed to overcome the effect of moist weather. Humidified Versus Non-humidified Sheds The results of the preliminary runs indicated that the first firing period for the 1932 crop needed to be considerably longer than was ordinarily required. As this firing is primarily designed to remove the great volume of water in the fresh tobacco, any factor that would hinder the removal of the water would make necessary firing either at higher temperatures or for longer periods at the usual temperatures. Such a factor was present in the thicker and heavier leaves caused by the dry weather at the time these leaves were developing. Temperatures much in excess of 90° F. are not advis- able, as the leaves on the lower tiers are apt to be scorched and there is danger of the production of green colors, "haying down," hence the necessity of prolonging the firing period. In this experiment, the effect of artificial humidification was studied in comparison with the lack of such equipment. From studies conducted in controlled chambers it seemed reasonable to expect benefits from humidification, as ordinary diurnal fluctuations give only a few hours per day during which the shed humidity is sufficiently high to permit absorption of water by the tobacco. Periods of high humidity and alternations of humidity in curing are desirable for several reasons: (i) The movement of solutes from the midrib to the leaf tissue is facilitated and fewer "fat stems" occur. (2) Areas of leaf tissue that cure at a slower rate do not become isolated and remain greenish in color. (3) The colors on the leaf are more uniform, as a better opportunity is pro- vided for diffusion of the water soluble pigments. (4) The chemi- cal changes incident to curing are encouraged, and the leaves are more thoroughly exhausted of excess carbohydrate and nitrogenous compounds. (5) Unless other factors intervene, thinner and more elastic leaves will be found. The experimental equipment used in the humidified shed was furnished by the American Moistening Company of Providence, R. I., and was designed to deliver 6 pounds of water per nozzle per hour, with eight nozzles to a bent, four just above the level of the fourth tier and four above the plate line. The check shed was filled with 13^ picking tobacco on July 25 and 26, and firing started at 6 p. m. of the latter date. The tem- perature was gradually increased during the first 16 hours, averag- ing 86° F. for this period, and was maintained slightly above 90° for 60 hours, followed by 20 hours at 87°. The weighted average for the entire period was 90°. The relative humidity rose sharply when the fires were lighted, and remained above 60 per cent for about 30 hours, dropped gradually to 40 per cent after 70 hours, 504 Connecticut Experiment Station Bulletin 350 and dropped sharply to nearly 30 per cent at the end of the firing period. Subsequent firing periods of 17 hours and 32 hours, both at average temperatures of 89°, were employed to remove excess moisture during rainy periods. The remaining record indicates only the ordinary diurnal fluctuations prevailing at this season. The record obtained from this shed shows most of the features of air conditions found in an average shed. The 30 hour high- humidity period during firing corresponds with the time during which the most rapid loss of free water occurred. Both the tem- perature and relative humidity were higher in the shed than outside, while the absolute humidity (grains of water per cubic foot of air) was 25 per cent higher. The period during which the leaves in this shed regained moisture was rather brief, roughly from midnight to sunrise, and averaged 7 hours. The maximum humidity of the day usually occurred immediately after sunrise, when the heat of the sun evaporated the condensed moisture on the leaves. The minimum day humidity averaged 44 per cent for the latter period of curing with nine days below 40 per cent. Thus it may be seen that the opportunity for optimum curing was rather limited, as only two periods of more than 12 hours with average humidities above 80 per cent occurred in 18 days. On July 26, filling of the humidified shed was commenced, but was interrupted by heavy rains on the 27th and 28th, and was not completed until the 29th. Firing was started at 6 p. m. of the last date and continued at an average temperature of 88.4° F. for 106 hours. The slightly lower temperature maintained in this shed made it necessary to continue the firing for 10 hours longer than in the check shed. As soon as the first firing was stopped, the moistening apparatus was turned on and a humidity of above 80 per cent maintained for 20 hours. Firing was then resumed for about an equal length of time at an average temperature of 87.5°. A period of humidifica- tion of 5 hours followed, and the afternoon of the following day another period of 4 hours. The humidity rose to more than 90 per cent during the night and in the afternoon of the next day the fires were lighted and allowed to taper off gradually for a total fir- ing of 16 hours at an average temperature of 87°. Subsequent to this firing, the tobacco was moistened for short periods at seven different times, as indicated by the chart, and fired for about 2 hours after a night of high humidity. As may be seen from the charts (Figure 138), it is possible to compare conditions in the two sheds at any one time, as the corre- sponding time periods are placed directly in line vertically. In general, the most conspicuous difference caused by humidification was a lengthening of the night period of high humidity. It was not possible with the equipment at hand in this shed to maintain Studies of Fuels for Curing 505 consistently a high humidity above a few bottom tiers when the outdoor humidity was low during the day. This is illustrated by the humidity graphs for the last two moistening periods, during neither of which it was possible to hold the moisture at a high level. As compared with the check shed, in which the average period with a humidity of 75 per cent or greater, was 7 hours per day, the humidified shed for the same period of time had 10 hours per day with humidity sufficient for tobacco to regain water. The mini- mum day humidity averaged 46 per cent for the final 16 days, with only 2 days below 40 per cent. Thus it may be seen that the tobacco in the humidified shed had considerably more opportunity to cure under favorable conditions. Considerable difference in the appearance of the tobacco during curing was noted. In the check shed, the colors were not as uni- form, some leaves being distinctly of the reddish brown type, others olive brown. Fleshy midribs, "fat stems," persisted for some time after their disappearance in the humidified shed. The distribution of color on the individual leaves was not as satisfactory, a consider- able spottiness prevailing on most of the leaves. Some staining of the leaf tissue from the midribs and secondary ribs was noted. These areas remained in a moister condition throughout the curing period, and the result was the development of a bright red color, which contrasted sharply with the color of the adjacent tissue. The most common colors in the darker areas of the leaves were medium to reddish brown, in the lighter areas yellowish to light reddish brown. In the humidified shed the colors were quite uniform in the whole lot of tobacco, all the leaves tending toward a medium olive brown to light olive brown. The distribution of color on the leaves was uniform, and staining from the midribs was practically absent. The leaves were thinner and more elastic, as might be expected with the increased proportion of favorable curing time and the greater opportunity for the chemical activities of curing. The length of time necessary for complete cure was considerably reduced in the humidified shed. Notwithstanding the fact that the curing period was started three days later in this shed, the tobacco was cured from a week to 10 days before the check shed and was taken down nearly 2 weeks before that in the check shed. After the tobacco had been sweated, notes were taken on colors, and sorting records obtained on samples from the humidified and check sheds. The most common color of the tobacco from the preliminary run on first picking was a light olive brown. Above the sixth tier in this shed the colors were not so bright and more leaves were mottled, indicating inferior curing conditions in the top of the shed. The tobacco from the preliminary run on i^^ picking was most commonly an olive brown to medium brown. From the check shed the predominant color was a medium brown, as was also the case in the humidified shed, but in the former the 506 Connecticut Experiment Station Bulletin 350 colors were slightly darker, and more leaves showed a "muddy" overcast. Above the plate line in both sheds, the colors were dis- tinctly yellower. It was observed by Mr. Kaffenburgh that after sweating, the tobacco cured in the humidified shed seemed to be further advanced and could be used on cigars with less ageing or "after-curing." In order to obtain an impartial opinion of the tobacco, the sam- ples were sorted in a commercial warehouse, the sorters not being informed of the treatments. The results are shown in Table 21. It is apparent from the grade indexes that the difference between the tobacco from the two sheds was negligible. Any differences which might have operated in favor of the humidified tobacco were almost entirely masked by the characteristics imparted to the tobacco by the climatic conditions during the growing season. These common characteristics were thickness of leaf, prominence of veins, darkness of colors, and lack of elasticity. Table 21. Sorting Record of Shed-Cured Shade Tobacco, 1932 Treatment Percentage of grades LL LL2 LC LV LV2 SLV YL AL Humidified Check 3-2 2.0 2.0 3-2 3-5 3-2 1.6 1-4 7-9 II. 6 10. 0 7.0 II. 8 10.6 4-3 5-3 Treatment Percentage of grades Grade LB sv SV2 LW XL K Br. index Humidified Check 6.3 30 6.0 8.8 14-7 12.9 5-7 8.7 14.0 12.0 71 7-9 1-9 2.4 1.392 1.360 Summary 1. An initial firing period varying from 48 to 106 hours at 90° F. was necessary with the 1932 crop. 2. Firing should be terminated when the leaves begin to show a tendency to become brittle, "huskiness," which condition corre- sponds with a relative humidity in the shed of 30 per cent in dry weather or about 45 per cent in moist weather. 3. The greater part of the original water content of the leaves is removed during the first 16-30 hours of firing. 4. The use of fires subsequent to the first firing period is neces- sary to prevent damage from pole sweat in damp periods, and may also be of benefit in hastening the rate of chemical activity in the late yellow or early brown stage. Studies of Fuels for Curing 507 5. The use of humidifying equipment in the shed permitted more rapid cure of the tobacco by supplying moisture for chemical activity. More uniform distribution of colors was found on the cured tobacco from the humidified shed, but the difference was not apparent after sweating. Fleshy midribs were not as common in the curing tobacco, but no difference in prominence of veins of the sweated tobacco was noted. 6. Tobacco did not absorb moisture from the air when the relative humidity was below 75 per cent. 7. Tobacco hung in the peaks of sheds was inferior to that placed in the body of the shed, and there is some question whether it is profitable to fill the peak of a shed. It was not possible, even in the humidified shed, to maintain a proper balance between tem- perature and relative humidity in the peak. 8. Due to the unfavorable climatic conditions under which the crop was grown, no definite information concerning the value of humidification was obtained. c o ^ 1 n p 0 W l-J ""»' 4i„ ■ '■--* *-' 'S— is5 a^r T V3 i ^fe-: 53s <= '^ jj^jsf ^;