Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Some effects of water table, pH, and ammonium and nitrate nitrogen upon the growth and composition of… Herath, Herath Mudiyanselage Edward 1967

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1967_A4 H47.pdf [ 5.37MB ]
JSON: 831-1.0104367.json
JSON-LD: 831-1.0104367-ld.json
RDF/XML (Pretty): 831-1.0104367-rdf.xml
RDF/JSON: 831-1.0104367-rdf.json
Turtle: 831-1.0104367-turtle.txt
N-Triples: 831-1.0104367-rdf-ntriples.txt
Original Record: 831-1.0104367-source.json
Full Text

Full Text

SOME EFFECTS OF WATER TABLE, pH, AND AMMONIUM AND NITRATE NITROGEN UPON THE GROWTH AND COMPOSITION OF HIGHBUSH BLUEBERRY by  HERATH MUDIYANSELAGE EDWARD HERATH B.Sc,  University of Poona, 1956  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN AGRICULTURE in the Division of Plant Science We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July,  1967  ii  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study.  I further agree that per-  mission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives.  It is understood that copying or publication  of this thesis for financial gain shall not be allowed without my written permission.  Division of Plant Science The University of British Columbia Vancouver 8, Canada July,  1967  i ABSTRACT Frequent drainage and aeration problems occur in blueberry plantings on acid peats (pH 3.0 to pH 4.2) of British Columbia during a part of the growing season. The effect of waterlogging, pH, and form of N were studied under greenhouse conditions.  Using one year old plants of Bluecrop  blueberry, a split plot design was employed with two water tables for main plots and a factorial combination of 4 pH levels and 3 levels each of ammonium and nitrate N (20, 40, and 60 lbs. N/acre).  An un-  fertilized check treatment was also included as a treatment. Growth records and leaf analysis showed that poor aeration under waterlogged conditions exhibited characters symptomatic of poor nutrition.  Sparse leaf growth, smaller leaves with severe yellowing  and premature leaf abscission were observed in the high water table treatments.  Leaf analysis revealed highly significant differences in  foliar N, P, K, Ca, Mg, and Fe levels. There was also a greater growth response to ammonium N and nitrate N.  Higher levels of nitrate N (40, and 60 lbs. N/acre) caused  severe leaf scorch. Although higher levels of ammonium N (40, and 60 lbs. N/acre) gave better growth response, growth was prolonged and f a l l leaf drop and wood maturation were delayed.  Plants receiving 60 lbs.  N/acre as ammonium N showed symptoms of dieback in the following spring. Although pH had very l i t t l e effect on leaf nutrient composition, growth appeared to be better at a pH level of around 4.2.  i i i (a) TABLE OF CONTENTS Page I.  INTRODUCTION  1  II.  LITERATURE REVIEW  4  III.  A.  Soil Reaction  4  B.  Source of Nitrogen  C.  Soil Aeration  9  D.  Leaf Analysis  10  8  MATERIALS AND METHODS  14  A.  Leaf Area Measurements  B.  Chemical Analysis  18  18  1. Method of Preparation of Samples for Chemical Analysis 20 2. Analytical methods C. IV.  Statistical Analysis  22 33  RESULTS AND DISCUSSION A.  36  Main Effects of Water Table  36  1. 2.  B.  General Observations 36 Effect of Water Table on Foliar Nutrient Composition 38 3. Discussion of Main Effects of Water Table 46 Influence of Water Table in Relation to Ammonium and Nitrate Nitrogen 49 1. 2. 3.  C.  Leaf Symptoms 49 Growth data 50 Flowering 52  Effect of Nitrate and Ammonium Fertilizer on Foliar Nutrient Composition 52 1. Discussion of N Nutrition and interactions 60  D.  Effect of pH on Growth 1. 2.  64  Effect of pH on Foliar Nutrient Composition 67 Discussion of Effect of pH on growth and Foliar Nutrient Composition 67  i i i (b) TABLE OF CONTENTS (continued) Page V.  SUMMARY  77  VI.  ACKNOWLEDGEMENTS  80  VII.  LITERATURE CITED  81  iv LIST OF TABLES Page Table Table Table Table  Table Table Table Table Table  1. 2. 3. 4.  5. 6. 7. 8. 9.  Table 10.  Foliar mineral element levels of highbush blueberry varieties  13  Analysis of variance for growth records and chemical analysis  16  Correlation coefficients for matrix involving leaf linear measurements . . . .  20  Effect of water table on mineral composition of foliage of Bluecrop blueberries  37  Effect of water table on growth of Bluecrop blueberries  37  Effect of source and rate of N f e r t i l i z e r on flowering  53  Effect of N source and rate on leaf nutrient element composition  56  Correlation coefficients for growth and nutrient elements of leaves . . . . . . .  57  Correlation coefficients for matrix involving leaf nutrients  58  Effect of pH on growth  66  Table 11. Effect of pH on flowering Table 12.  Effect of pH on foliar nutrient composition  68 71  V  LIST OF FIGURES Page Figure Figure Figure Figure Figure  1. 2, 3. 4. 5.  Prediction of leaf area from linear measurements  19  Drifting baseline due to insufficient lamp conditioning  25  Determination of calcium by atomic absorption  27  Determination of magnesium by atomic absorption . . . . . . . . . . . . . . . .  28  Determination of iron by atomic absorption  29  Figure  6.  Example of a blocked atomizer . . . . .  30  Figure  7.  Measurement of absorption lines . . . .  34  Figure  8.  Effect of water table and date on  Figure  9.  Effect of water table on foliar P . . .  40  Figure 10.  Effect of water table onffoliar K . . .  41  Figure 11.  Effect of water table on foliar Ca . . .  42  Figure 12.  Effect of water table on foliar Mg . . .  43  Figure 13.  Effect of water table on foliar Fe . . .  44  Figure 14.  Effect of water table, source and rate of N on shoot growth  51  Effect of source and rate of N f e r t i l i z e r on leaf N content  54  Figure 16.  Effect of pH on shoot length  65  Figure 17.  Effect of water table on pH on leaf  Figure 15.  number . . . . . . « • • • • . . . . * .  69  Figure 18.  Effect of pH and water table on leaf area  70  Figure 19.  Nitrate toxicity  74  I.  INTRODUCTION  The highbush blueberry, (Vaccinium corymbosum L.) i s cultivated under a wide range of s o i l and climatic conditions i n the south, west, and northeastern regions of the North American continent. It i s commercially popular as a horticultural crop i n eastern North Carolina, southern New England, western New York, southern Michigan and western Oregon and Washington. The highbush blueberry is also cultivated on a commercial scale in the southeastern region of Canada and to a limited extent on the western coastal region of British Columbia. Climatically, successful cultivation is limited to the regions mentioned above due to the specific chilling requirements of the highbush blueberry. From the edaphic aspect, the most important factor i n the s o i l appears to be the pH. Due to peculiar nutrient requirements of the highbush blueberry, ecological adaptation appears to depend to a large measure on i t s s o i l requirements.  If the natural flora in a  particular region where i t is to be grown contains related plants of the Ericaceae, i t i s a good indication that the highbush blueberry w i l l succeed.  In order to adjust the s o i l pH to the optimum range,  corrective measures have been adopted in many areas either by the addition of dolomitic lime to raise the pH, or by the addition of ground S to lower pH. Mineral soils as well as soils of organic origin can be utilized for growing blueberries.  Most commercial highbush blueberry plantings i n the lower Fraser Valley delta region have been established on acid peat and muck s o i l s . to pH 4.2.  The pH in these soils ranges from pH 3.0  Wherever the surface layer of peat has been exploit-  ed the remaining layer of decomposed organic material may range from a l i t t l e over a foot in thickness to a few feet overlying sand or sandy clay. Due to the nature of the flat terrain, the delta region has a very high water table inundating blueberry fields for periods sufficient to create waterlogged conditions. Although drainage f a c i l i t i e s have been provided in these fields, the water table often goes above the 14 to 30 inch level which i s considered to be ideal for blueberries (20). The nutrient supply of these peats does not seem to be sufficient for successful cropping.  Under such acid conditions  microbial activity and mineralization of the organic N would be very slow.  Poor drainage and low s o i l pH would reduce the uptake  and availability of P, basic cations, and some trace elements like Cu and Mo. A preliminary survey of blueberry leaf nutrient status was carried out in the summer of 1965 to obtain some guidelines which would serve as a basis for the initiation of a comprehensive program of research leading to the establishment of c r i t i c a l nutrient levels and to establish criteria for making f e r t i l i z e r recommendations for highbush blueberry.  A closely related objective of that prelim-  inary survey was to evolve a reliable precedure for collecting leaf  - 3 samples.  Since the survey revealed possible "hidden hunger"  levels for N and P according to the estimated deficiency levels presented i n Table (1) (Cain and Eck, 14), the present experiment was designed to study the effects of two sources of N at different pH levels maintained under waterlogged and free-drained conditions.  II. A.  LITERATURE REVIEW  Soil Reaction Early work of Chandler (15) showed increase in yields  with application of lime at the rate of 8 tons per acre to lowbush blueberry fields.  The high dose of lime used did not seem  to have adverse effects on the crop. Although these i n i t i a l studies were conducted on lowbush blueberries, the response to liming, as indicated by better growth and higher yield, prompted further study extending to the highbush blueberry.  Cain and  Galletta (12) found the range of pH 4.5 to pH 4.8 to be best suit ed to successful commercial production of highbush blueberries. Johnston (34) reported that pH was more c r i t i c a l than any other s o i l factor.  Merril (39) found that pH 3.2 or lower was too acid  often causing detrimental effects to plants, perhaps because at such low pH levels most nutrients were limiting.  These results  were in accord with those obtained by Stene, (44);who submitted that true acidity was not the ultimate criterion blueberry survival was dependent upon, but rather the level of nutrition they were able to derive from such a nutrient environment.  Even at  higher pH levels blueberries thrived, provided the balance of supply of nutrients was maintained.  Kramer and Schrader (36)  suggested that the reason why blueberries thrived in acid soils was due to their low cation requirements. They found as a rule most acid soils were low in exchangeable cations, and the blueberry absorbed excessive quantities of anions other than the  - 5 phosphate. They also theorized that Fe deficiency might result in the plant i f available Fe in the reduced form were oxidized by the anions.  Bailey et a l ( 7 ) in their investigations observed  similar trends which indicated further that foliar levels of P, K, Ca, and Mg were lower in blueberry than in most fruit crops. Using Rubel highbush blueberry plants, Doelhert and Shive (25) demonstrated that optimum growth was obtained from treatments high in N and low in K. A drop in the pH of these nutrient solutions also indicated that the ammonium ion was taken up more than the nitrate ion.  This trend was observed more at  higher pH levels. Cain (11) also suggested that ammonium sulphate gave better results in soils above pH 5 . 5 . Willis and Carreo (45) produced chlorosis in rice with several forms of n i t rate N. They suggested that an unassimilable ion caused precipitation of Fe within the plant resulting in chlorosis. In experiments conducted by Cain (11), varying Ca content in the nutrient solution, typical Fe deficiency symptoms in high Ca treatments were obtained where the source of N was calcium nitrate.  It was also found that as Ca in the foliage increased, N,  P, Mg and Fe decreased.  From these studies he made the following  generalizations: 1.  The supply of ammonium N determined the ultimate levels of foliar N and Fe.  2.  Plants which were chlorotic at high Ca levels, had a high Fe content.  - 63.  Iron deficiency symptoms were not necessarily related to s o i l pH, Ca content or Fe content in the foliage, since healthy plants had more Ca and less Fe than those showing acute chlorosis. The experiments of Cain (11) were conducted at pH 5.5 and  i t was apparent that ammonium N was influencing both Fe metabolism and growth.  In later experiments Cain (13) showed that Fe deficien-  cy symptoms were closely linked with the pH of the leaf tissue. Neutralization of free organic acids would mark the point at which these symptoms appear. Considerable speculation as to the role of the free basic amino acid arginine arose from this work. At the point where deficiency symptoms appeared, the level of arginine increased with a corresponding decrease i n protein N. This indicated a very close relationship between N, Fe, and leaf pH. A chlorotic condition that appeared i n the midsummer of 1966 in most blueberry fields i n the lower mainland of British Columbia was similar to that described by Bailey (4). These symptoms were more evident i n heavy bearing plantings.  I l j i n (32) found similar trends and  in addition he noticed that the chlorotic condition was associated with a higher content of organic acids.  Cain (13) also found that a higher  s o i l :pH induced acgreater uptake of basic cations.  These findings re-  sulted in more organic acids getting neutralized, giving the higher leaf pH values he observed. He further found a large proportion of citrate in chlorotic leaves.  These conditions would decrease the solu-  b i l i t y of Fe and inactivate the enzymes that depend on Fe as a co-factor.  - 7 It would thus appear that organic acid metabolism, N metabolism and chlorophyll synthesis are closely related to the Fe dependent enzyme action as suggested by Cain.  Oertli (40) also re-  ported similar interdependent relationships between N metabolism and soil pH in connection with the role of Fe in the plant.  Bail-  ey and Everson (6) demonstrated chlorotic symptoms, and possibly Fe deficiency by adding rates of lime ranging from 5-40 crock. 6.4.  grams per  After nearly 14 months the pH values ranged from 4.2 to  This experiment was apparently done with mineral s o i l . By varying growth media and pH, Hall ejt al (29) showed  that the lowbush blueberry, like the highbush varieties, grows best in a range of pH 4 to pH 5.  Although Bailey (5) pointed out that  the addition of peat partially alleviated the effect of lime induced chlorosis, Hall et al (29) failed to show this effect by an interaction of pH with media. There was, however, a high negative correlation between stem length and s o i l pH, a condition which could not be shown under field conditions. Since pH values of peats in British Columbia are much lower than the optimum range reported in the above investigations, the effect of different rates of liming on the growth of the highbush blueberry grown on low pH peat soils, needs further study. A l though i t has not been conclusively established, i t was observed during the preliminary survey of 1965 that plantings which were receiving regular lime applications appeared to have better growth. Some growers also claimed that plots regularly limed tended to give  higher yields. Soil pH levels in such mature plantings were slightly higher than in newly established fields. B  «  Source of Nitrogen The effect of nitrate and ammonium N have been studied i n  a series of experiments by Colgrove and Roberts (18) on the growth and development of chlorosis in the Azalea, which like the highbush blueberry,is also a member of the family Ericaceae.  Citing the  findings of Cain (11) on the preferential uptake of ammonium N and its influence on Fe metabolism i n the blueberry, they worked on chlorosis by making a comparative study of ammonium and nitrate N under different environmental conditions, of light intensity and pH in the presence and absence of chloride and sulphate ions in the nutrient solution.  These experiments they conducted were in sand culture.  Their findings demonstrated an antagonism of ammonium ions with basic cations, a greater uptake of the former reducing uptake of the latter. This brought about a reduction of plant tissue pH. Nitrate on the other hand increased base absorption, and Fe was "inactivated" in the plant as a result of the higher tissue pH. An influence of nitrate on absorption of bases like Ca, Mg, and K has also been demonstrated by Sideris and Young, in pineapple (43), and by Holley, Pickett and Dulen (31), i n cotton. From the diverse studies cited i t appears that close relationships may exist between s o i l pH, form of N desired and common symptoms like chlorosis. It is imperative to clarify the combination of ammonium and nitrate N with differing s o i l reactions and amounts of N that would  give the most favourable response. C.  Soil Aeration In most areas on the lower mainland where the highbush  blueberry is grown, peat soils are often poorly drained.  The water  table remains a few inches from the surface during a greater part of the autumn, winter and spring. At times flooding may occur in prolonged wet weather.  Even though many farms are drained, rain water  percolates very slowly in peat, with the result that fields remain flooded long enough to cause adverse conditions in the root zone. If plants grow under these conditions for too long, the blueberry develops a matty root system on the surface and around the collar, thus raising above the fluctuating water table. This phenomenon is popularly called " h i l l i n g " by growers.*  By this unique adaptation i t  appears that older plants are able to survive even i f the deeper roots are subject to a low redox potential brought about by poor aeration, i f complete inundation does not cause irreparable damage. There is no doubt, however, that growth can be retarded under such conditions (14). In mineral soils, Render and Brightwell (35) have reported that poorly drained conditions have detrimental effects on blueberry plant growth. The highbush blueberry does not tolerate standing water for long periods during the growing season. Building of mounds and ridges on lowland soils has been a common practice to promote root growth above the reach  Eaton, G;W., and H.M.E. Herath, Unpublished survey of leaf nutrient status in highbush blueberry plantings in British Columbia 1966.  - 10  -  of the water table. Waterlogged plants on mineral soils were found to be less productive than plants in well-drained locations.  The  slow growth and occasional death of newly set plants in locations frequently subjected to waterlogging, could well be due to the i n ability to generate sufficient roots near the surface.  The high  water retentive capacity of peat may also create a situation where air spaces are displaced by water.  Peat can retain as much as twenty  times i t s weight in water. The root system of the highbush blueberry is devoid of root; hairs (19).  This may be an ecological adaptation.  The much branched  and fine root system is in intimate contact with the s o i l solution and the need for a specialized absorbing region becomes redundant.  No  studies have so far been carried out to determine the extent to which highbush blueberry plants can stand excess water in peat without seriously affecting normal physiological processes and depressing growth. D.  Leaf Analysis as a Diagnostic Tool in Blueberry Fertilizer Programs Many techniques are available today for the study of plant  nutrition in agricultural research.  The use of a specific organ or  tissue warrants justification. Leaf analysis has been successfully used in f e r t i l i z e r programs for perennial crops as a supplement to the more conventional technique of s o i l analysis (9, 42).  It helps in studying more closely the  nutrient requirements of plants in relation to higher yields and quality production.  Evaluation of the nutrient requirements of a crop is based  on the assumption that the leaf acts as an indicator tissue and reflects  - l i the utilization of essential elements at any time during the l i f e of the plant((42). There are various reasons for selecting the leaf as an indicator tissue.  Primarily the leaf acts as the site for the synthesis  of raw materials out of which various plant tissues are composed. A l though other tissues in the plant may be suitable for detecting a nutritional disorder, the leaf tissue has been widely used due to i t s accessibility, sensitivity (for most elements), and homogeneity.  For  physiological and practical considerations, the leaf tissue has therefore been accepted as the most convenient plant organ to study nutritional disorders.  Variability in sampling can also be minimized i f mor-  phologically homologous tissue of comparable physiological maturity is used for analysis (42). The level of nutrients in the leaf, as in most tissues of the plant, appears to have a high correlation with metabolic indices such as growth and yield.  The sensitivity of the leaf is such that when the vig-  our of the plant is impaired by any abnormality in the nutrient or physical environment, i t registers a characteristic symptom. In the f i e l d , however, difficulties arise in diagnosing certain symptoms due to the multiplicity of factors that interact and affect growth.  It is at this  stage that leaf analysis proves valuable as a diagnostic tool in detecting hidden or latent symptoms that cannot be found out by s o i l analysis. A s o i l test is s t i l l the primary means by which f e r t i l i z e r requirements are determined for crops (42).  The method of s o i l analysis i f used as  the sole means of diagnosis has i t s drawbacks too.  The availability of  - 12 s o i l nutrients or the extent to which a perennial fruit crop could derive from a s o i l various nutrients at different depths cannot be accurately estimated from a conventional s o i l test.  Since the ul'-i. «ta  timate effect is on the plant, nutritional disorders may go undected in a s o i l test which could easily be found by tissue analysis. It could also be beneficial in assessing responses to a f e r t i l i z e r or differentiating a mineral deficiency symptom caused by a nutrient disorder from that of a non-nutrient cause (9). The leaf analysis technique has been effectively used to establish f e r t i l i z e r programs for the highbush blueberry in other areas on the North American continent.  Table I gives the consoli-  dated data obtained by various workers.  The results of a sampling  survey conducted in the lower Fraser Valley of British Columbia i n 1965 have also been included for comparison.*  Median shoot leaves  and basal leaves from shoots subtending a fruit cluster were sampled from two varieties during the months of August and September.  The  results of this survey for the varieties Dixi and Rancocas, and also for the variety Bluecrop from the work of 1966, included in Table I comprise the grand mean for each element from the statistical analysis.  Eaton, G.W., and H.M.Ev Herath, Unpublished survey of leaf nutrient status i n commercial highbush blueberry plantings in British Columbia 1966.  Table I. Foliar Mineral Element levels of Highbush Blueberry varieties Source  Variety  Bailey et a l  Rubel  Mikkelsen et^ a l  % N  % K  % P  % Ca  % Mg  Fe ppm  0.13-0.18  0.53-0.68  0.2-0.4  0.12-0.22  Jersey  0.18-0.32  0.40-0.87  0.3-0.5  0.14-0.21  60-84  Mikkelsen et a l  Jersey  0.28-0.32  1.2  0.2-0.3  0.14-0.19  70-  Mikkelsen et^ a l  Jersey  0.26-0.27  0.80-1.2  0.18-0.3  0.22-0.24  60-70  1.9-2.0  Ballinger et a l  2.0-  0.16  0.53  0.74  0.28  150  Ballinger et_ a l  1.5  0.07  0.40  0.30  0.09  60  Tukey et a l  2.2  0.22  0.74  0.47  0.24  290  0.14-0.18  0.42-0.56  0.2-0.3  0.09-0.25  Popenoe  Rancocas  Cain  Jersey  1.9-2.6  0.11-0.16  1.2-1.8  0.4-1.2  0.10-0.15  72-170  Cain  Rubel  1.8-2.4  0.07-0.09  0.43-0.78  0.5-0.7  0.13-0.17  60-85  Cain  Rubel  1.1-1.2  0.04-0.07  0.61-0.97  0.33-0.6  0.09-0.24  67-144  Cain  Rubel  1.4-2.3  0.07-0.09  0.37-0.43  0.5-0.7  0.03-0.08  Eaton and Herath  2  Rancocas  1.36  0.09  0.34  0.28  61  Eaton and Herath  2  Dixi  1.57  0.09  0.43  0.31  65  Eaton and Herath  2  Bluecrop  1.56  0.09  0.28  0.21  0.43  95  All  1.8  0.07  0.40  0.30  0.08  60  Estimated deficiency  Information obtained from Childers' Fruit Nutrition cited by Cain and Eck (14) Unpublished.  - 14 III.  MATERIALS AND METHODS  The plant materials used i n this experiment were one year old rooted cuttings of the Bluecrop variety of highbush blueberry obtained from a commercial nursery.  The rooted cuttings were sel-  ected from a large population for uniformity of size of top and root system. As far as possible roots were washed clean of extraneous matter and the tops of plants consisted of a single stem with two to three short side shoots.  The Bluecrop variety was selected  for these studies due to i t s increasing popularity as a mid-season variety in British Columbia as well as elsewhere.  Summer growth  had already commenced when the plants were set i n the greenhouse. Peat for the growth medium was obtained from the Northern Peat Company, Richmond, British Columbia.  The pH of the peat before  adjustment varied from pH 3.0 to pH 3,5.  In order to ascertain the  amount of hydrated lime required to adjust pH to predetermined  levels,  weighed quantities of the peat were incubated with known amounts of lime.  The wet peat thus treated was gently agitated for seventy two  hours i n a mechanical shaker and the pH recorded.  At the end of one  week the readings were taken again and the resultant values were plotted against amounts of lime added, and this buffer curve was used to estimate the required amounts of lime to adjust the pH to three levels above the original level of 3.4. Although i t was the intention to adjust levels increasing by one unit, the ultimate values obtained were close to 3.4, 4.3, 5.2, and 6.0. Eight-inch glazed crocks with drainage holes in the side were prepared in the following manner. Treatments receiving a high water  - 15 table were sealed with tight fitting rubber stoppers through which a bent glass "sight tube" was inserted to serve as an indicator of the water table in the crock. With this device i t was possible to regulate the water table i n the crock to any desired level along the height of the peat column and create a r t i f i c i a l l y the conditions of a waterlogged  soil.  Two blocks i n the experiment received this treatment.  The  other two blocks were maintained under free drained conditions giving ample water to keep the peat moist a l l the time.  These four blocks  were randomized on two greenhouse benches. Each block comprised 28 crocks with seven treatments at each pH level.  The seven treatments  were as follows. Three levels equivalent to 20, 40, and 60 lbs. N per acre were given in the form of ammonium sulphate and sodium n i t rate.  In addition to these six treatments an unfertilized check was  included for each pH level. block.  Treatments were randomized within each  The degrees of freedom for the analysis of variance for growth  measurements and chemical analysis of foliage are presented i n Table 2. Randomization within blocks was changed periodically to evenly d i s t r i bute bias due to position. Plants were set three weeks after liming. The f e r t i l i z e r treatments were given when the plants were well established. The water level treatments were maintained throughout the growing season.  The  two blocks receiving the high water table treatment had water three inches from the surface. Over-irrigation of the treatments maintained  Table 2. Analysis of variance for growth records and chemical analysis Shoot Number  Shoot Length  df  Source  Flower Clusters FL/CL CL/P1  Leaf Area  Leaf Number  Source  df  Source  df  Source  Chemical Analysis Single Combined Date Date df  Source  df  Source df  Source  df  Source  W  11  W  1  W  1  W  1  W  1  W  1  W  1  W  1  df  B/W  2  B/W  2  B/W  2  B/W  2  B/W  2  B/W  2  pH-W  3  pH-W  3  PH  3  pH  3  pH  3  PH  3  PH  3  pH  3  Tr  6  Tr  6  pH-W  3  pH-W  3  pH-W  3  pH-W  3  pH-W  3  pH-W  3  pH-Tr  18  Tr  6  Tr  6  Tr  6  Tr  6  Tr  6  Tr  6  Error  139  Tr-W  6  Tr-W  6  Tr-W  6  Tr-W  6  Tr-W  6  Tr-W  6  Date  2  18  D-pH  6 12  Tr-pH  18  Tr-pH  .18  Tr-pH  18  Tr-pH  18  Tr-pH  18  Tr-pH  pH-Tr  18  Error(A)27  Tr-pH-W 18  Tr-pH-W 18  Tr-pH-W 18  Tr-pH-W 18  Tr-pH-w'• 18 Tr-pH-W 18  D-Tr  Error  54  Error  54  Error  54  Error  54  Residual54 Error 336  Residual54 Error 112  Error(B)56  TOTAL  111  TOTAL  111  TOTAL  111  TOTAL  111  TOTAL 447  TOTAL 223  W - water  B/W - Blocks within water table  Tr - Treatment  TOTAL  167  D - Date  TOTAL 167  - 17 under free drained conditions was avoided to minimize leaching of N f e r t i l i z e r from the crocks. A month after application of f e r t i l i z e r , the f i r s t leaf samples were taken. As i n field experiments, leaves were sampled from the mid-shoot region.  In order to minimize the  effect of leaf removal, only about four leaves were removed from each crock.  Subsequent samplings carried out at monthly inter-  vals were obtained in similar fashion using leaves of comparable physiological age to reduce leaf variability due to leaf age differences.  Leaf disc sampling which i s at times followed in  greenhouse experiments of this nature in which leaf removal may set back growth, was not found necessary here as there were sufficient leaves per plant, after removal of these few leaves for mineral analysis,for normal growth.  A composite sample from the  two blocks of each treatment was used for chemical analysis. Since the experiment commenced in July the three sampling dates were in August, September and October. Growth continued into the end of November when a r t i f i c i a l heating i n the greenhouse was discontinued to allow plants to winter properly.  No casualties were  observed during the course of the experiment, although severe stunting and lesions occurred due to certain treatments.. Light intensity appeared to be sufficient i n the greenhouse although slightly higher day temperatures accelerated growth considerably in comparison with field grown plants.  During the day the mean  maximum temperature was always a few degrees higher in the greenhouse than outside.  - 18 Growth records were taken at the end of the growing season. The records included total leaf number, mean leaf area, shoot number and total shoot length. Flowering records were taken i n the spring of 1967. Soil pH readings were recorded periodically during the growing season. A.  Leaf Area Measurements Mean leaf area was measured by selecting the f i r s t four  mature leaves in the terminal region of a shoot on each plant. Shoots were selected at random. The method for the estimation of leaf area was as follows: In order to obtain the closest correlation of a linear measurement to actual leaf area, a regression equation was derived. A sample of forty leaves were sampled at random ranging from the smallest to the largest.  Ozalid paper was then used to obtain leaf impressions.  The greatest length and width of each leaf was then measured after which the paper impressions were weighed and leaf area calculated.  Re-  gression of these measurements i n a l l possible combinations with actual area was carried out.  The best correlation (r-0,999) was obtained bet-  ween area and the product of length X width.  The linear relationship  obtained i s presented i n Figure 1 and Table 3 and indicates a stronger association than any reported by Ackley et_ a l (1) or by Jain and Misra (33) on Ricinus communis.  The linear measurements and the regression  coefficient for the actual area was calculated from leaf samples of the experiment. B. Chemical Analysis The leaf samples were subjected to chemical analysis.  Total  N was analysed by the semi-micro kjeldahl procedure, P was determined  .•I  :3LQ  m  i  ~0'  L E N G T H  6  0  W I D T H :  I  - 20 Table 3. Correlation coefficients for matrix involving leaf linear measurements and leaf area.  Area  Length  Width  Length X Width  0.9495  0.9766  0.9991  -0.1675  0.9794  0.9775  0.0164  0.9503  -0.1587  Length Width  Length/Width  -0.1675  Length X Width  colorimetrically by the phospho-molybdate method of Dickman and Bray (22).  Atomic absorption spectrophotometry was used for the determin-  ation of Ca, Mg, and Fe.  Using the same instrument (Model 140 EEL  Atomic Absorption spectrophotometer) with an emmission adaptor, K was analysed by flame emmission. 1.  Method of Preparation of Samples for Mineral Analysis  The wet digestion method of Chapman and Pratt (16) was used for the preparation of the mineral extracts. When leaf samples were brought i n from the f i e l d , they were immediately placed in a 70°C forced air oven for a minimum of 72 hours. Before grinding they were transferred to a 105°C oven for seven hours. A porcelain mortar grinder was then used for grinding the samples until they were fine enough to pass a 20 mesh seive. sample was seven minutes.  Approximate grinding time for each  No metal contamination occurred as no mov-  ing parts inside this grinder are made of any metal that might con-  - 21 taminate the samples.  The dried and ground samples were then  stored in glass containers and placed in desiccators until the mineral extracts were made. Since the drying technique used was thorough, no attempt was made to make additional corrections for moisture.  From the results of the work done i n 1965 i t was  apparent that this technique was very reliable.-'-  No duplicate  determinations were carried out on any of the 1966-1967 samples because the duplicate determinations done in the previous year agreed within one percent error, and because only the means of observations were of interest. From the tissue samples thus prepared, close to one gram of sample was weighed for wet digestion.  A digestion mix-  ture consisting of n i t r i c , sulphuric and perchloric acids in the ratio of 750 ml. concentrated n i t r i c acid, 150 ml. of concentrated sulphuric acid and 300 ml. of 607« perchloric acid was prepared according to the method described by Chapman and Pratt (16) and 10 ml. of this mixture was added to each sample of tissue. This was heated gently for a few minutes and the temperature raised unt i l the fumes of nitrogen dioxide f i r s t disappeared, followed by perchloric and sulphuric acids.  Taking care not to allow complete  drying, the solution volume was reduced to about 2-5 ml. (P would  Eaton, G.W., and H.M.E. Herath, Unpublished survey of leaf nutrient status in commercial highbush blueberry plantings in British Columbia 1966.  - 22 be lost i f taken to complete dryness). After digestion was complete, hot  d i s t i l l e d water was added to the beaker and the extract filtered  through acid washed f i l t e r paper into a volumetric flask.  The f i l t e r  paper was washed with more hot d i s t i l l e d water to. ensure the removal of a l l the extract and the solution was made to 100 ml. volume with d i s t i l l e d water.  A l l the acids used i n the extraction procedure were  redistilled from a glass s t i l l , since minor elements were to be analysed.  The extracts thus prepared were transferred to "Nalgene" con-  tainers, using aliquots of this when required for various determinations. 2.  Analytical methods  Atomic absorption spectrophotometry Analyses for Ca, Mg, and Fe were done by Atomic absorption. The Atomic absorption method uses the "ground state" atoms which do not acquire sufficient energy from the burner flame to emit light.  Since  there is a greater proportion of these so-called "ground state" atoms the  absorption technique is more sensitive and accurate than emission  spectrophotometry, especially for metals occurring in minute quantities as in extracts of plants.  The absorption of light at the specific  "resonance" wavelength for each metal is proportional to i t s concentration in the sample solution.  A monochromator wavelength selector is  used in combination with a hollow cathode source lamp to generate the 'resonance' wavelength for each element to be determined (27). Zero absorbance i s obtained when the light source of hollow cathode lamp passes unobstructed through the monochromator from where an electrical signal equivalent to the amount of absorption is amplified by the photomultiplier.  When a sample is introduced the meter reading drops corres-  - 23 ponding to the amount of light absorbed. The advantage in this method lies in the fact that sensitivity could be increased or decreased depending on the range in which a particular element occurs in a sample.  If the sample concentration is too low,  special organic solvents or extraction methods could be used. If the range of standards selected shows a linear relationship with absorbance, any unknown sample could be directly read off by comparison with the curve obtained from the standards, provided the sample concentration falls within the range of standards used.  If the relationship is not linear, a log transform-  ation facilitates accurate reading. Instrumental parameters When the Atomic absorption method was used various instrumental parameters were considered, including the following: 1.  Finding the optimum analytical absorption line for each element analysed.  2.  Best s l i t width for the monochromator.  3.  Optimum lamp current for the hollow cathode lamp.  4.  Various optimum flame parameters depending on whether absorption is best in a high or low flame or whether i t gives maximum  absorption in an oxidizing or reducing flame.  The sen-  s i t i v i t y of the instrument can thus be adjusted by controlling fuel supply to alter flame height and quality.  Other  parameters such as the rate of sample uptake and range of absorption were given due consideration as per methods described by A.O.A.C. techniques on atomic absorption spectro-  - 24 photometry for minor element determinations (2). The f a c i l i t y provided for burner rotation afforded additional f l e x i b i l i t y in the use of the instrument. N» serious interferences were observed during the analysis of any of the elements to justify the use of alternate methods. There were no interferences from phosphate or sulphate when Mg was analysed. Even for Ca the phosphate interference at 40 ppm was negligible.  Dis-  cernible interference was only shown at lower concentrations around 10 ppm.  Since most of the samples had values above 30 ppm, La or Sr  were not used in the standards or samples. An Evans Electroselenium Ltd. atomic absorption spectrophotometer model 140 was used with a Texas Instruments Servo recorder model FS01WGA. A distinct advantage in the use of such an arrangement was the greater speed at which samples could be analysed. The chart also made i t easy to check with the standards periodically in order to see whether the instrument was steady throughout the run. A l l readings were corrected from the readings obtained from the blanks. At approximately every 50th sample, sets of standards were taken for the computation of a least squares fitted curve. Standards as well as samples were run only after the i n strument stabilized without any baseline d r i f t .  This usually occurs  due to poor adjustment of the monochromator, a block in the atomizer or when the hollow cathode lamp has not been conditioned enough, ure  (Fig-  2). Although i t is claimed by the manufacturers that the lamp has  to be heated f«r only 30 minutes i t was found necessary to heat i t for a minimum of three hours prior to use. When Ca was being analysed, the  -  25  -  Fig2DRIFTING BASELINE DUE TO INSUFFICIENT LAMP CONDITIONING  CATHODE  - 26 burner was rotated to give the desired sensitivity and range since higher concentrations were present in the samples.  Since the re-  solution was poor at high concentrations the range selected for Mg was from 0.5 up to 6 ppm.  Therefore a l l samples were diluted 2500  times before they were used on the instrument. Interpretation of Data from Chart Sets of readings for standards and for the elements Ca, Mg, and Fe as read on chart paper, are shown in Figures 3-5.  It was  observed from these charts that the absorbance lines for a particular sample could easily be read with the minimum of error i f the baseline is steady.  During the course of a run, a l i t t l e solid impurity in the  samples could block the atomizer assembly.  This would result in a jag-  ged line at the base as well as at the peak (Figure 6). As soon as this was observed, the atomizer unit was cleaned by inserting a tiny wire through the aspirator. boiling the unit in water.  The atomizer could also be cleaned by This was usually carried out as a routine  preventive measure at the end of the day when the instrument has been used f»r a long time.  To cite an example to justify this procedure,  i t was found that the Ca in the samples even at low dilution of 1 in 100 caused a white encrustation of calcium carbonate a l l along the aspirator assembly.if the instrument had been used for a few days without proper cleaning. A better cleaning procedure was to dip the atomizer for about five minutes in a very weak solution of hydrochloric acid and subsequently boiled in water for at least ten minutes.  In  this way the atomizer can be cleaned and kept ready for the next run.  - 27 -  - 29 *  —  j  1  -  -  " - :  (J  —  -i.  —  i- - -  •l  —  1 -  "i '  r  r  > io"  -  •  —  -  r e a ..a.  ~T'  :  ------  — -  i  E  o  a a •A - f • |  1  .  Fig J  :  f  — -  s  n a- a rt  -  a _a  -  &—  -- o'  -  —— z  '  -—" -  ;  :  ~  -1  • •  _  1  ATOMIC  -  -4— •  1  § .1  BY  K  p  . . . k j s . .  J^-  i-J  IRON  P "  -  1  OF  ~  —  - -  -1-  DETERMINATION  \-  1  »  _ — -  . _ E  - '  i  . E  1 j  1 i  —  1  Sinndard 5  — --  —  ;  -  — f~  ' '  -  --- -  • -  i  -  !  '-- - -  ft  ii •  i —  1  —  --—  -1  •  ,  ABSORPTION  i  <!  li 1  V 1  A  a 30 w  - 31 A poorly functioning atomizer would give a lesser reading due to smaller amounts atomizing into the flame.  In general about 10%  of the spray from the atomizer passes through the expansion chamber in the form of minute droplets that finally vaporize into the . flame. The larger droplets collect at the bottom of the expansion chamber and leave via the water trap outlet. The rate of atomization and the physical properties of the sample to be aspirated play a v i t a l role i n the production of maximum efficiency in absorption. From preliminary runs i t was apparent that the diluted extract showed no P interference at the range in which Ca determinations were made. The same effect could be obtained by the use of an efficient atomizer.  The size of droplets is dependent on the fineness of the  spray given out by the atomizer.  In the case of interference by  compounds such as calcium phosphate, i f there is a greater production of droplets that vaporize sufficient Ca atoms into the flame the effect of interference is considerably diminished.  According to E l -  well and Gidley (27) the exact mechanism of interference and suppression is not fully understood although the use of many organic solvents or the addition of chlorides of Sr, La, Fe, and Sc have been tried out with some degree of success.  The fact remains however, that a simple  adjustment and the proper care and use of a fine atomizer unit i n the aspirator assembly would give ample signal strength for routine plant analysis work. The r e l i a b i l i t y of the method of atomic absorption has been verified and the fact that results are reproducible when the same  - 32 instrumental parameters are employed lends support for i t s use in plant analysis work. Chart data for standards of the work done in 1966 and 1967 agreed very closely when the adjustment for each element was carried out in the same manner. Finally, the causes of interference affecting both absorption and emission are summarized below as given by Elwell and Gidley (27). 1.  The effect of different physical characteristics of solutions on atomizer efficiency.  2.  Closely linked to the above condition the differential rates of vaporization due to variability of solutions used. Both these are influenced by the viscosity of the solution and its ability to form droplets easily.  3.  The proportion of atomized solution that enters the flame to be separated into excited or ionized atoms and ground state atoms would largely depend on the flame characteristics. The interferences mentioned above are practical aspects  of atomic absorption spectrophotometry one has to consider when this method is employed for plant analysis work. If the same set of conditions are given when an element is analysed a number of times, the results thus obtained are comparable.  During the course  of the present investigations a l l instrument parameters were kept as uniform as i t was practically possible. Each peak on the chart as seen in the Figures 3-5 represents the concentration of the element being analysed for a particular sample. The height of this line is measured from the base to  33 the widest portion of the peak. L i t t l e specks of suspended particles in the samples temporarily blocking the atomizer would at times tend to increase and decrease signals momenta r i l y , resulting in a jagged peak. Figure 7 demonstrates the way to measure the base and peak of any given absorption line. Since the variability in measurements from chart data can easily reduce the efficiency of an experiment, a uniform technique of reading chart data was found necessary.  In the case of elements  like Fe where the peaks are small and differences between samples are harder to detect, a slight error in reading may well exceed the magnitude of the residual standard deviation for that set of data. Measurement of these lines were either taken in inches or by counting the number of lines that span across the chart paper.  These values were utilized in two ways for statistical  analysis.  Originally the values for the standard curves were cal-  culated manually for a least squares f i t , and the standard curve thus obtained was used to interpolate the sample values after they were corrected for the blank readings. This method was later streamlined by punching the readings for the standards as well as for the samples directly on computer data cards. The concentration of each sample was then calculated on the computer using the preliminary data deck, and a modified fortran linear regression program. C.  Statistical Analysis Analysis of variance was carried out on a l l the data  collected.  The subdivisions of degrees of freedom for the various  * 34 *  ..J  i  i  ~&9-7  i  METHOD" OF MEfcSMWNGr  PEMC  i  SKHPLE CONCENTRATIONS:  - 35 analyses are presented in Table 2.  Significance of differences among  means were tested by Duncan's New Multiple range test as described by L i (38).  Significant results of single effects as well as interact-  ions are presented in charts and tables with the incorporation of Duncan's Multiple range test values.  Unless otherwise stated, a l l  results reported were significant at the 57. level or lower.  - 36 IV. A  »  RESULTS  Main Effects of Water Table 1.  General Observations  Of the variables studied in this investigation, a high water table had a profound effect on certain aspects of growth and nutrition.  As the plants responded to the effect of the change in  the root environment due to a high water table, distinct discoloration of leaves appeared. The original pale color developed p i g m e n t - . ation similar to multiple deficiency symptoms. These symptoms became more pronounced as the season advanced. Apart from these qualitative visual ratings between the two water table treatments no chlorophyll estimations were attempted due to lack of sufficient leaf tissue for analysis. Growth retardation appeared to manifest i t s e l f in smaller, b r i t t l e and deformed leaves.  This difference in leaf size was  ficant only at the 5% level, (Table 5).  signi-  It is also apparent from the  data presented in this table on other growth records that the two water regimes did not show significant differences.  Visual observations how-  ever showed marked differences in the general appearance of the plants. Because the testing term used for water table was blocks/water table in the model used for the analysis of variance, and because the variation between blocks was  large, any greater variance due to the water table  alone could have easily been obscured.  The large variation among  plants within each treatment was mainly due to the difficulties confronted in maintaining a steady and continuously uniform water table as the rates of water loss from plant to plant differed widely.  The  Table 4.  Effect of water table on mineral composition of foliage of Bluecrop Blueberry  Month  Water table  N  P  K  Ca  August  High Low  1.68 d 1.99 b  0.094 c 0.129 b  0.308 d 0.355 c  0.193 b 0.213 b  0.488 b 0.520 ab  100 ab 99 ab  September  High Low  1.68 d 2.25 a  0.121 b 0.221 a  0.356 c 0.614 a  0.163 c 0.214 b  0.326 d 0.529 ab  85 be 106 a  October  High Low  1.76 c 2.23 a  0.132 b 0.199 a  0.378 b 0.615 a  0.204 b 0.242 a  0.403 c 0.588 a  Mg  1  Fe  ,'81 c 90 b  Means i n the same column followed by the same letter are not significantly different (Duncan's New Multiple Range Test - 5% level) Mean of 28 plants.  Table 5.  Effect of water table on growth of Bluecrop Blueberry  Water table High Low  Leaf number  Leaf area  N.S 95.98 112.23  * 24.74 32.82  Shoot number N.S 7.35 8.71  * Significant difference,between water tables at,P. = .05 N.S. - not significant  Shoot length N.S 142.8 178.9  - 38 same problem arose in the free drained crocks. In certain cases even regular watering failed to prevent irreversible drying out of the peat during the height of the summer. 2. Effect of Water table on Foliar Nutrient Composition Highly significant differences appeared in the uptake of a l l elements studied.  The data presented in Table 4 and in Figures 8-13  show that N, P, K, Ca and Mg were substantially decreased by a high water table in comparison with plants maintained under free drained conditions.  The effect on foliar iron was not so profound at the  beginning of the season, but showed a highly significant difference in October when leaves on plants maintained under waterlogged conditions exhibited symptoms of yellowing and premature aging. Foliar N was decreased by the presence of a high water table. Highly significant reduction in leaf N content was obtained on a l l three dates of sampling (Table 4 and Figure 8). The yellowing effect brought about by the water saturated s o i l condition could be seen well during the middle of the season and i t became more apparent as the leaves aged.  Senescence and leaf abscission were hastened under these  adverse conditions.  Since the uptake of a l l the elements was affected,  i t would appear that the basic metabolic activities were slowed considerably.  One effect of N f e r t i l i z e r application was the significant  increase i n level of foliar N over that of the unfertilized control. The N level was highest for the second date i n the low water table treatment; while N levels of the leaves decreased i n the low water table treatments as the season advanced, they tended to increase slightly in the waterlogged medium.  The date X water interaction  EFFECT  OF WATER TABLE O N FOLIAR N  0.  CO  AUG  Fag. 8.  SEP  OCT  SEASONAL CHANGE OF FOLIAR N LEVELS UNDER HIGH WATER TABLE AND FREE-DRAINED CONDITIONS. Means sharing the same letter are not significantly different (D.N.M.R.T. 5%)  EFFECT OF WATER TABLE O N FOLIAR p  0-30  0-25  i  0-20  low  a.  o  *  0-15  <  L  O ooo 0-05  „  B6  J  0-00  AUG Fig. 9.  SEP  1966  i  L  OCT  SEASONAL CHANGE IN FOLIAR P AS INFLUENCED BY HIGH WATER TABLEL AND;iFREE-rDBAINED CONDITIONS  '  EFFECT OF WATER TABLE O N FOLIAR K 0-7  0-6  1  "•*••«••  Low  1-0-5 vs  a.  s/0-4  I  .High  < ^0-3  o  0-2 B6 1966 I  AUG  Fig. 10.  SEP  i  I  OCT  SEASONAL CHANGE IN FOLIAR K AS INFLUENCED BY HIGH WATER TABLE AND FREE-DRAINED CONDITIONS.  EFFECT OF WATER TABLE O N FOLIAR Ca 0-3  _  <—0-2  _  U 0. a  CM  <  o-i  00  AUG  Fig. 1 1 .  SEP  OCT  SEASONAL CHANGE IN FOLIAR Ca AS INFLUENCED BY HIGH WATER TABLE AND FREE-DRAINED CONDITIONS.  EFFECT  Fig. 12.  OF  WATER TABLE O N  AUG  SEP  FOLIAR M g  OCT  SEASONAL CHANGE IN FOLIAR Mg AS INFLUENCED BY HIGH WATER TABLES AND FREE-DRAINED CONDITIONS.  EFFECT  Fig. 13.  OF W A T E R TABLE IRON  ON  FOLIAR  SEASONAL CHANGE IN FOLIAR Fe LEVEL AS INFLUENCED BY HIGH WATER TABLE AND FREE-DRAINED CONDITIONS.  - 45 (Figure 8) was significant (P-0.05). The seasonal trend for foliar P was similar to that of N. Differences between the two water regimes were highly significant for a l l three dates. The interaction of date X water table was highly significant.  Plants receiving a regular supply of water  without any waterlogging took up more P i n midseason than when they were waterlogged.  This trend is normally observed in crops that are  well supplied with phosphate.  No significant differences were ob-  tained between dates when plants were maintained under high water tab le ,cond i t ions. There was a highly significant increase of leaf K under free-drained conditions, the level remaining high even at the end of the season.  Under waterlogged conditions the level of foliar K  remained low even i n October.  The differences between the two water  regimes were more marked with progressive maturing of leaves as compared with the negligible increase in early summer. Uptake of K thus appears to have been impeded when the root system was subjected to oxygen stress. The fluctuation of Ca levels was more pronounced under waterlogged conditions than when plants were maintained under freedrained conditions.  A low water table gave higher Ca values in the  foliage on a l l three dates of sampling, the difference being highly significant (P«0.01) i n the months of September and October.  In Aug-  ust, however, leaf Ca showed no significant difference between the two water treatments. In September, the leaf Ca level dropped considerably in the high water table treatment, and as growth slowed  - 46 down towards the end of season, the level increased. There was a depression in foliar Mg when the water table was high, the decrease being very evident and highly significant in both September and October. The date X water table interaction was highly significant.  There was no difference due to water regimes in  August. Whilst leaf Mg gradually increased under well aerated conditions, there appeared to be hardly any change in Mg level of leaves in the poorly drained conditions since the i n i t i a l depression brought about by poorly drained conditions.  The behaviour of leaf Mg in the  two water table treatments was very similar to that of Ca.  Mg levels  were, however, higher than Ca levels in a l l plants maintained under high as well as low water treatments. Significant differences i n foliar Fe levels showed up late in the season.  Premature leaf aging due to the waterlogged condition  seemed to be closely related to the low N level and resultant yellowing of leaves.  In September the high water table treatments register-  ed a drop in foliar Fe while an opposite trend was observed in plants under free-drained conditions. 3.  Discussion of main effects of water table,  Results of this experiment indicate a distinctive sensitivity of the highbush blueberry to waterlogged conditions.  Although growth  indices studied did not show significant differences in a l l cases, there is ample evidence to conclude that waterlogged peats do not provide the ideal conditions for growth or for maintaining a proper balance of nutrients in the highbush blueberry.  The primary cause may be the lack of  - 47 proper aeration and consequent interference with normal functioning of the root system. The lower levels of leaf N, P, K, Ca, Mg, and Fe observed in the waterlogged treatments do not necessarily indicate that the waterlogged peat was unable to supply these elements in sufficient quantity.  Since both sets of crocks received similar  f e r t i l i z e r treatments, i t is reasonable to suspect that poor aeration impeded normal physiological functions of the roots.  Analogous con-  clusions have been reached by Buttery et_ al (10), Gore and Urquhart (28) and by Labanauskas et_ al_ (37).  They concluded that growth re-  duction under waterlogged conditions could not be directly attributed to external limitations of the nutrient environment. When citrus plants were grown under oxygen stress as a result of excess water treatments, Labanauskas et a l (37) reported that total leaf N, C l , Zn, Cu, and Fe decreased significantly.  They also  found that total P, K, Ca, Mg, Mn, and B were not affected.  In the  study made by Gore and Urquhart (28) on Eriophorum vaginatum and Molinia caerulea, i t was found that waterlogging did not affect P levels but caused severe N deficiency. They also reported that waterlogging resulted in low redox potentials in the peat medium. In the present work on highbush blueberry, the results on foliar nutrient levels under waterlogged conditions do not seem to be in accord with the above findings, except in the case of N and Fe.  As presented i n  the results, highly significant differences were also obtained for leaf P, K, Ca, and Mg. The severity of symptoms of the effect of a highbush blueberry may have been due to the prolonged wet condition affecting the root system which was confined to such a small volume  - 48 of peat in the crocks.  An additional factor partly responsible for  these striking differences may have been the prevention of any i n ward diffusion of air from the sides or the base of the crocks.  By  and large these conditions may prevail under field conditions, long enough to affect growth and performance of the blueberry.  At this  stage i t is only possible to speculate as to the physiological nature and cause of these observed differences. In order to make a c r i t i c a l study of the damage to the root system, further investigations are necessary.  Estimation of respiration rates, root regenerative capa-  city with constant and fluctuating water tables were not within the scope of this experiment. Although the highbush blueberry is supposed to tolerate a high degree of water excess, i t is obvious from these findings that better growth could be obtained by regulating the water supply.  The  shallow rooted nature of the crop clearly indicates that survival in such environments depend to a large measure on the fine network of roots the plants have closer to the surface.  Profuse root growth on  the surface and around the collar of the waterlogged plants in the experiment showed the so-called "hilling effect" as seen in the f i e l d . There were clear indications that the a r t i f i c i a l l y created high water table induced a greater rooting propensity at the surface where i t was beyond the reach of the water table, unlike in the well aerated peat where the root distribution was more uniform.  Field grown blueberries  when growing under prolonged submerged conditions in the root zone, appear to exhibit characters symptomatic of multiple nutrient deficiencies.  This i s often seen in poorly drained locations where plants  - 49 also show i n addition to these deficiency symptoms, a typical "staghorn" appearance, poorly colored leaves and severe stunting even though such fields have been supplied with f e r t i l i z e r s . B. The Influence of Water Table i n Relation to Ammonium and Nitrate Nutrition 1. Leaf Symptoms: One week after f e r t i l i z e r treatments were administered the foliage registered symptoms of marginal necrosis and leaf abscission which were assumed to be toxicity symptoms. The condition was more pronounced i n the nitrate than i n the ammonium treatments. At 20 lbs. N per acre, both forms of N showed no "toxic" effects. per acre, only nitrate treatments were affected.  At 40 lbs. N  At 60 lbs. N per  acre, "toxicity" was observed on plants with both sources of N, the symptoms being more severe in the nitrate treatment. "Toxicity" symptoms varied slightly under the two water regimes.  Plants receiving a high water table showed less marginal  necrosis, but leaf abscission occurred i n the treatment receiving the equivalent of 60 lbs. N as nitrate.  Under free-drained conditions tox-  icity was shown in the form of severe necrosis, appearing f i r s t on the tender shoot tips and progressing down to the mid-shoot region. Most plants recovered sufficiently after a few weeks and normal healthy growth was observed in the new foliage, except those plants receiving the 60 lbs. N per acre as nitrate, under both systems of water management. Leaf growth on these plants was very sparse and the plants never fully recovered.  - 50 2.  Growth data: Using total shoot length as an index of growth, s i g n i f i -  cant interactions were obtained between N source and water table (Figure 14). In the high water table treatment a significant growth response was obtained over the check when ammonium N was applied at 40 and 60 lbs. per acre of N. No significant differences were obtained between the two forms of N at 20, 40, or 60 lbs. N when the water table was high. Under free-drained conditions overall growth was greater at the low f e r t i l i z e r levels, in contrast to the trend in the high water treatments.  Depression of growth in the check was less severe  than when the root zone was under oxygen stress.  At 20 lbs. per acre  of applied N both forms of f e r t i l i z e r gave a better response than the nitrate treatment of 60 lbs. N application.  Although the difference  in shoot growth between the two forms of N at the applied level of 20 lbs. per acre were not significantly different, the ammonium treatment was also significantly different from the nitrate treatment of 40 lbs. N under similar free-drained conditions.  The differences in  shoot length between the check and the three f e r t i l i z e r treatments or between source of N at each level were not significant in the low water treatments. Comparing the two water table treatments, poor growth was obtained in the check and the 20 lbs. N per acre level when the water table was high, but far better growth occurred at 20 lbs. N level when plants were grown under free-drained conditions.  No significant differences  were obtained at 40 and 60 lbs. N per acre between the two water treatments. Neither leaf number, mean leaf area nor shoot number  EFFECT  OF  WATER  TABLE,SOURCE  SHOOT  High  water  APPLIED  Fig. 14.  OF HIGHBUSH  OF  N  O N  BLUEBERRY  table  Jcheck ^nitrate j ammonium  0  GROWTH  A N D RATE  20 N  •o  w  o  a  40  60  I N  0 LBS PER  20  ACRE  A COMPARISON OF THE TWO FORMS OF NITROGEN NUTRITION ON SHOOT GROWTH OF THE 'BLUECROP' VARIETY OF HIGHBUSH BLUEBERRY GROWN UNDER WATERLOGGED AND FREE-DRAINED CONDITIONS. MEAN SHOOT LENGTHS SHARING THE SAME LETTER ARE NOT SIGNIFICANTLY DIFFERENT (D.N.M.R.T. 57.)  - 52 were affected by the forms of N. At the end of the growing season i t was observed that high N treatments delayed leaf f a l l and in a few cases older leaves remained on the plants throughout the winter. 3. Flowering: The f e r t i l i z e r effects on flowering were highly significant. The number of flower clusters per plant and the number of flowers per cluster (mean of two clusters taken from the third position from the terminal bud) were determined and the flowering records analysed statistically.  The check was significantly lower in flower cluster number  than a l l other f e r t i l i z e r treatments except the one receiving 60 lbs/ac N as nitrate (Table 6). The data on flower number per cluster showed similar trends. In the ammonium treatments, blooming was earlier than in the nitrate treatments by as much as four days. C.  Effect of Nitrate and Ammonium Fertilizer on Foliar Nutrient Composition: N, P, K, Ca, and Mg levels of the foliage were affected by  the form of N as well as by the different rates of application. In August, foliar N levels were not affected by the treatments. This was perhaps due to the plants not being fully established in the crocks. Significant differences did appear however, by the middle of September. The analysis of results on this date indicated a significant response in leaf N to both forms of applied N (Figure 15). There were no significant differences between the two sources at 20, 40, or 60 lbs. per acre of N application.  Differences were, however, significant between  levels of applied N. The check had significantly lower leaf N levels  - 53 -  Table 6. Effect of source and rate of N-fertilizer on flowering Treatment Check  lbs. N/acre  Clusters per plant  2  Flowers per Cluster^  0  3.69 c  2.69 c  3  20  13.19 ab  5.69 ab  4  20  18.50 a  6.91 a  3  40  12.88 ab  5.10 ab  4  40  18.50 a  5.87 ab  N0 NH  N0 NH  N0  3  60  4  60  NH  6.44 be 15.00 a  Standard error = 2.3114  1  4.53 be 6.13 ab Standard error = 0.7346  1. Means within a measurement followed by the same letter do not differ significantly (P=0.05) 2. Mean of 16 plants 3.  Mean value of two clusters per plant was taken for statistical analysis and the mean of 16 such values represent each mean in the table.  EFFECT O F S O U R C E  A N D RATE O F N  FERTILIZER O N FOLIAR  N  CONTENT  B6 1966  o  to LBS  Fig. 15.  4o PERACRE  6 0 N  THE RELATIONSHIP BETWEEN APPLIED N IN THE FORM OF NITRATE AND AMMONIUM NITROGEN ON FOLIAR N CONTENT OF BLUECROP BLUEBERRIES. 1  1  - 55 than ammonium N treatments of 20, 40, and 60 lbs. applications.  On  the other hand, the check was only significantly lower than the nitrate treatment of 60 lbs. N per acre.  Both forms of N at 60 lbs. N  application gave significantly more leaf N than at other levels of applied N, (Table 7), in September.  Since this sampling date was in  mid-season, i t was considered the most crucial phase of growth when foliar nutrients could occur in limiting concentrations. The correlation coefficients in matrix form presented in Table 8, show s i g n i f i cant correlations obtained between leaf nutrients analysed, and the different growth indices measured.  In September, leaf N was s i g n i f i -  cantly correlated with shoot length, leaf number, flower cluster number and shoot number. Although leaf N levels at the end of season are not of immediate importance, substantial amounts present at this time may imply a steady supply for the spring flush. the October sampling was carried out.  Therefore analysis of  The check plants in October had  significantly lower N than a l l other f e r t i l i z e r applications.  There was  a progressive decrease of leaf N in the check from the beginning to the end of season.  As in September, there was no significant  between the two forms of N at each level of application.  difference There was no  significant correlation of October leaf N levels with any of the growth indices studied. Leaf P levels followed trends similar to those of N.  The  overall leaf P values showed a highly significant positive correlation with leaf N in September (Table 9). Judging from the level of leaf P obtained at this time of year, there appeared to be a beneficial effect  Table 7.  Month  E f f e c t of N source and rate on leaf nutrient element composition , sampling dates Check  Aug % Sept N Oct  1.69 1.57 d 1.43 c  Aug 7. Sept P Oct  l b s . N/Acre  20 N0  3  NH4  40 l b s .N/Acre  60 l b s .N/Acre  N0  N0  3  NH4  1.75 cd 2.05 ab  1.83 1.89 be 1.98 ab  1.84 1.79 1.82 bed; 2.05 b 1.87 b 2.14 ab  0.11 0.09 b 0.07 c  0.11 0.19 a 0.15 b  0.11 0.19 a 0.19 a  0.11 0.16 ab 0.15 b  Aug 7. Sept K.Oct  0.31 0.28 b 0.27 b  0.34 0.55 a 0.50 a  0.31 0.57 a 0.55 a  Aug 7. Sept Ca Oct  0.21 0.22 0.21 0.19 0.22 ab 0.19 b  Aug 7. Sept Mg Oct  0.50 0.48 0.37 cd 0.33 d 0.43 cd 0.37 d  Aug ppm Sept Fe Oct  *  94 111 80  2.01  129 112 81  3  on three  Standard Error  NH4  1.79 2.34 a 2.26 a  1.86 2.35 a 2.25 a  0.08 0.08 1.10  0.11 0.19 a 0.20 a  0.11 0.18 a 0.20 a  0.10  0.005 N.S,  0.16 ab 0.19 a  0.02 0.01  0.36 0.44 ab 0.46 b  0.34 0.43 ab 0.51 a  0.35 0.57 a 0.64 a  0.32 0.55 0.55 a  0.02 N.S. 0.06 • 0.06  0.19 0.17 0.21 ab  0.18 0.15 0.20 ab  0.20 0.19 0.23 ab  0.21 0.21 0.23 ab  0.20 0.20 0.28 a  0.02 0.02 0.01  N.S. N.S.  0.50 0.51 ab 0.52 be  0.52 0.33 d 0.44 cd  0.54 0.47 0.51 abc 0.39 bed 0.48 cd 0.67 a  0.51 0.55 a 0.56 ab  0.04 0.05 0.04  N.S.  94 88 87  95 92 96  11 12 3  N.S. N.S. N.S.  97 88 84  98 87 87  94 92 85  N.S.  Means within a single date sharing the same l e t t e r are not s i g n i f i c a n t l y d i f f e r e n t (Duncan's New Multiple Range Test - 57.). Treatment means f o r one element on a single date without l e t t e r s , were not s i g n i f i c a n t l y d i f f e r e n t on Analysis of variance (P=0.05), and were not tested with Duncan's New Multiple Range Test. The standard errors are given for means on a l l dates.  - 57: -  Table 8. Significant correlation coefficients for growth indices and nutrient elements o leaves 1 Foliar N  P  K  Ca  0.53  0.48  0.53  0.34  Leaf number  0.33  0.38  0.52  Flower clusters  0.28  Shoot number  0.44  Shoot length Leaf area (L X W)  0.27  0.32  0.43  * Only significant correlations are presented. C r i t i c a l r at 57. = 0.26 Critical r at 17. = 0.34  with leaf analyses in September  - 58 -  Table 9.  Correlation coefficients for matrix involving leaf nutrients 1  Relatidnship between leaf nutrient elements in September  N N —  P  K  Ca  Mg  0.36  0.50  —  -0.35  0.40  0.56  -0.34  --  0.37  -0.26  —  -0.35  P  --  —  K  --  --  Ca  —  —  —  Only significant correlations are presented. C r i t i c a l r at 5% =0.26 C r i t i c a l r at 1% = 0.34  Fe  0.37  0.77  - 59 of N applications onifoliar P, a clear demonstration of the existence of an N-P relationship at these leaf concentrations observed.  The P  level of check plants were not significantly different from the nitrate treatment of 40 lbs. N and the ammonium treatment of 60 lbs. N per acre. A l l other treatments were significantly higher than the check.  Signifi-  cant correlations were observed for P values in September with shoot length, leaf number and shoot number, (Table 8).  In October, P levels  were highly correlated with shoot number. Leaf K levels in August did not show any significant differences.  This trend was also observed in the case of N and P. The check  plants i n September were significantly lower in foliar K than when plants were given N f e r t i l i z e r at the rate of 20 and 60 lbs. per acre. This difference was, however, not significant at the 40 lbs. N rate of application.  Leaf K values were positively correlated with leaf area,  shoot length and shoot number in August; with shoot length, leaf number, flower cluster number and shoot number in September and with the same indices of growth in October.  When leaves were analysed in October, K  level of check plants was significantly lower than a l l other applied N levels.  In either September or October the differences between the two  sources of N were not significant at any level of N application. In spite of the liming treatments in this experiment, leaf Ca values were l i t t l e affected.  No significant effects were observed, ex-  cept in October where the 60 lbs. N application i n the ammonium form gave a significantly higher leaf Ca value than plants grown with 20 lbs. nitrate N.  - 60 As in the other major elements already discussed, leaf Mg did not show significant differences  in August.  The analysis  of results for September gave significant differences between the two sources at each level of application.  in leaf Mg A l l plants  maintained at 20, 40 and 60 lbs. N per acre with ammonium as their source of N gave significantly higher leaf Mg than plants grown with equivalent levels of nitrate N. The leaf Mg values remained in much the same manner in October save for an overall increase at the higher levels of N application.  In September foliar Mg was  negatively correlated with leaf N, P, K and Ca (Table 9).  This may  well be the manifestation of ion antagonism at this crucial period in the nutrition of the blueberry. In this experiment leaf Fe content did not seem to be affected by the source or the level of N used. This may have been due to the fact that the leaves were analysed for the total Fe fraction, which does not give a true indication of the active form of Fe in the leaves. Fe levels.  Cain (11) found chlorotic plants to have high  No significant correlations were obtained between Fe  content and any of the factors of growth studied.  Leaf Fe was, how-  ever, significantly correlated with leaf P and Ca in September. 1,  Discussion of N nutrition and interactions  The observed increase in the levels of leaf N, when higher levels of ammonium fertilizers were given, lend support to the theory that the principal form of N preferred by the highbush blueberry is the ammonium ion.  The results of Cain (11) also  indicated  - 61 that this was the case as far as highbush blueberries were concerned. The absence of sufficient ammonium ions in calcareous soils could probably be the reason why blueberries fare poorly i f adequate quantities of ammonium fertilizers are not supplied.  Although no men-  tion is made of the exact pH range, Cain and Eck (14) report that in New Jersey, Michigan and North Carolina, nitrate N fertilization i n highbush blueberry plantings is widely practiced on acid soils with good results. Fertilizer mixtures used by blueberry growers in the lower mainland of British Columbia commonly contain the three sources of N, namely Urea, ammonium sulphate, and ammonium nitrate in varying proportions in combination or singly. By resorting to this practice they feel that a steady supply is assured.  A similar recommendation  has been made by Doelhert (24) for very sandy soils.  It i s , however,  a matter of conjecture, whether the s o i l conditions would ensure an adequate supply of N to meet the growing demands of a heavy bearing crop like the highbush blueberry. The movement of ammonium ions has been studied by Townsend and cited by DeLong (21).  Experiments on the movement of the ammon-  ium ion when added in the form of a fertilizer to the s o i l in highbush blueberry fields in Eastern Canada, showed that penetration was slow due to i t s retention by the s o i l colloidal complex closer to the surface.  The ammonium ion, however, is weakly held by the s o i l c o l l -  oids and almost with the same intensity as the K ion. But both these ions are held with less tenacity than the H ion in an acid s o i l .  Due  - 62 to its loose attachment with s o i l colloids, heavy rain or a constantly fluctuating water table, could easily leach the ammonium ion. These experiments also showed that leaching occurred as much as 3 inches with an application of one inch of water. The leaching of ammonium ions in a mineral s o i l could be due to the difficulty of replacement of Al ions when ammonium N is added to the s o i l .  These conditions discussed  here are perhaps too drastic to expect in a highly organic environment where the matric potential of the s o i l solution is not only governed by hydrostatic, gravitational and absorptive components but also by the more important entities of an organic s o i l like osmotic concentration at the low pH as obtaining in these peat bogs. Although the acid nature of the peats would promote a preponderance of free ammonium ions, other factors could limit the availability of ammonium N. Since denitrification is mainly brought about byjthe microflora, poor aeration as a result of a high water table would result in low microbial activity.  The optimum pH range for  activity of such.organisms is not ideal in these peat soils. Mycorrhizal associations of blueberry with certain basidiomycetes have been reported by Coville (19) who  claimed  that in acid soils the fungi involved in this association transform ithe organic N into available forms while Rayner (41) described a process which brought about the fixation of atmospheric N by the mycorrhiza.  It has since been shown by Addoms and  - 63 Mounce (3) that this amount of N was insufficient to sustain the plant.  There is l i t t l e doubt that nitrogen nutrition of  the blueberry is dependent on the s o i l pH under natural conditions.  On the other hand, by supplying the plants with the  desired source of N, successful cultivation of the blueberry could be done at any pH level provided a proper nutrient balance is maintained in respect of the other essential elements in the s o i l (44). The increase of leaf P with the increase of N applications mentioned in the results of this greenhouse investigation was also observed in a contemporary field investigation on two year old "Bluecrop" blueberries.  At certain concentra-  tions of each element, i t has been shown by the use of multiple curvilinear regression methods by Dumenil (26) that the c r i t i c a l N-P concentration is not a point nor a narrow range of values, but a wide range of concentrations.  Within the limitations of  this experiment and the results from the f i e l d , i t can be speculated that at a c r i t i c a l point of one element or both, there appears to be a highly positive correlation when the elements are in a highly mobile state in mid-season. The cation antagonism of Mg has also been clearly demonstrated in this investigation. The inability of plants to obtain sufficient N at low N levels when the plants were waterlogged and the lack of different i a l responses to water treatments at higher levels of N may  be  - 64 due to two reasons. Under waterlogged conditions the uptake of N may be impeded by malfunction of the absorption mechanism of the roots purely due to lack of proper aeration.  It may also  have been due to dilution by excess water in the crock.  No  serious leaching was possible from the high water table treatments since i t was a closed system as far as the movement of water was concerned. D.  Effect of pH on Growth The four pH levels used in this experiment were arbit-  rarily selected to find out whether pH influenced growth and also to find out the effect of liming in relation to water table and form and level of N supply.  Results indicated that shoot  development was better above pH 4.3.  There was no significant  difference in total shoot length between pH 5.2 and pH 6.0. Plants maintained at pH 3.4 were significantly lower in shoot length than grown at pH 5.2 and pH 6.0 but they did not differ significantly from plants grown at pH 4.3 (Figure 16). did not induce an accumulation of Ca in the foliage.  Liming  The high-  bush blueberry appears to be frugal in i t s Ca requirements.  The  effect of liming on growth, evident from these results (Table 10), was apparently indirect. but not so markedly.  Liming also affected other growth indices,  More shoots were produced at pH 6.0 than at  any other pH level. Soil pH influenced flowering in the following season. Both flower cluster number per plant and the number of flowers per  EFFECT  O F pHO N  3-4  4-3  SHOOT  5-2  LENGTH  6 0  PH  Fig. 16.  THE INFLUENCE OF pH ON SHOOT GROWTH OF 'BLUECROP BLUEBERRIES GROWN IN PEAT, 1  (Means sharing the same letter are not significantly D.N.M.R.T. 5%)  different.  - 66 -  Table 10. Effect of pH on Growth and Flowering pH  Shoot Length  Shoot Number  Leaf Number^  Leaf Area^  3.4  139.4 b  7.3 b  87.3  25.7  7.3 b  4.2 b  4.3  159.3 ab  7.7 b  117.9  29.5  14.8 a  6.5 a  5.2  173.3 a  8.2 ab  105.9  30.3  16.7 a  5.9 ab  6.0  171.6 a  9.0 a  105.2  11.5 ab  4.4 b  7.47  1.21  Standard Error 8.80  0.41  2.96  Flower Clusters  FL/CL  Means within a measurement sharing the same letter are not significantly different. (P =0.05) 2 No main effects were significant but interactions with pH shown in Figures 17 and 18 were significant.  - 67 cluster were affected. Maximum flowering appeared to be around pH 4.3 and pH 5.2. A lower pH suppressed flower bud development (Table 11). This effect on flowering is probably only a reflection of the effect upon shoot growth. Although leaf area and leaf number measurements showed no significant main effects, the pH X water table interactions were significant (Figures 17 and 18). Comparing the two Water tables at each pH level, significantly higher leaf numbers were obtained in the low water table treatments only at pH 4.3. A l l other differences were not significant. A study of leaf area of treatments at the four pH levels showed that water table and pH affected leaf size (Figure 18). With a high water table pH 3.4 produced significantly smaller leaves than plants grown at pH 6.0. No significant effect of pH level was found when plants were grown under free-drained conditions. 1.  Effect of pH on Foliar Nutrient Composition  Except for some significant differences of leaf N and K in October and leaf K in August, the pH levels studied did not significantly affect foliar nutrient levels (Table 12). Higher pH values i n these instances tended to increase the uptake of N and K. 2. Discussion of effect of pH on growth and foliar nutrient Composition Since the blueberry has low cation requirements as cited earlier (36), and since the plants used in this experiment were nonbearing, there was probably not much growth stress on the plants and l i t t l e demand for nutrients from the peat.  Furthermore, peat i s  - 68 -  Table 11. Effect of pH (Liming) on flowering  Soil pH  Clusters per Plant  z  Flowers per Cluster  3.4  7.32 b  4.21 b  4.3  14.82 a  6.50 a  5.2  16.71 a  5.88 ab  6.0  11.54 ab  4.45 b  Standard error • 7.4714  Standard error = 1.2014  Means within a measurement followed by the same letter do not differ significantly (P=0.05) Mean of 56 plants Mean of 56 plants and two clusters per crock  EFFECT  OF WATER  TABLE  O N  5-2  6 0  LEAF  NUMBER  3-4  4-3 PH  Fig. 17.  LEAF NUMBER AS INFLUENCED BY THE INTERACTION OF WATER TABLE AND pH (LIMING). (Means sharing the same letter are not significantly different) (D.N.M.R.T. 57.)  EFFECT  OF pH AREA  ON LEAF  o <  Fig. 18.  LEAF AREA AS INFLUENCED BY THE INTERACTION OF WATER TABLE AND pH. (Means sharing the same letter are not significantly different) (D.N.M.R.T. 5%)  Table 12. The Effect of pH on Foliar Nutrient CompositionsInitial pH values Element  Month  3.4  4.3  ^  5.2  6.0  S-  y  (  7cN  Aug Sept Oct  7.P  Aug Sept Oct  7oK  Aug Sept Oct  %Ca  Aug Sept Oct  0.20 0.21 0.21  0.20 0.18 0.23  0.20 0.19 0.23  0.21 0.18 0.22  0.01 0.02 0.01  7<Mg  Aug Sept Oct  0.52 0.46 0.54  0.50 0.37 0.52  0.49 0.45 0.45  0.51 0.44 0.48  0.03 0.04 0.03  Aug Sept Oct  93 113 85  90 86 87  ppm Fe  *  *  * *  1.71 1.80 1.80 b  1.91 1.97 1.99 b  1.84 1.87 0.06 2.12 1.97 0.08 2.12 a 2.08 a 0.07  0.10 0.20 0.16  0.12 0.16 0.18  0.11 0.15 0.16  0.11 0.15 0.16  0.00 0.02 0.01  0.28 b 0.34 a 0.35 a 0.36 a 0.02 0.47 0.46 0.46 0.55 0.05 0.40b 0.49 ab 0.46 b 0.64a 0.05  99 93 85 .  116 92 85  9.0 9.3 2.7  Means showing significant differences among pH levels according to the analysis of variance (P=0.05) for each date were tested with Duncan's new multiple range test.  1 Means sharing the same letter are not significant (P=0.05)  - 72 supposed to have a high buffering capacity and any release of additional nutrients for the plants may have been slow. A l though growth in general was affected by pH, the distribution of minerals in the foliage was not significantly affected. Soil acidity does however, contribute indirectly in combination with other factors as seen from the results discussed earlier. The importance of s o i l acidity has been stressed by Coville (19). Johnstono(34) specifically cites the examples of commercial plantings grown on mineral soils that are used for cultivation of this crop.  However, in plantings in the lower Fraser valley  which are mainly established on peat soils, the problem appears to be different in that the pH is often too low.  In most of the  areas of the lower mainland, a newly developed peat bog would have a pH of about 3,0 and in older plantings i t may go up to pH 4.7.  The rise in pH in these mature plantings would probably  be due to the use of dolomitic limestone on the peat bogs as well as due to the increased microbial activity. Although leaf mineral status was not much affected by the pH treatments given in this experiment, s o i l pH appear to play a v i t a l role in the performance of the highbush blueberry. Merril (39) found that low pH levels were detrimental to blueberries.  This may perhaps be due to the preponderance of free H  ions competing for ion exchange sites or the solubility of a greater amount of heavy metal ions in the s o i l solution. Ammonium and H ions have a profound influence on the lower cation ab-  - 73 sorption due to high concentration of these ions on the root film.  Clark (17) found that ammonium ions reduced the uptake  of bases in strawberry resulting in lower organic acids in the leaf tissue.  A similar condition was observed by Cain (13) i n  highbush blueberry.  Colgrove and Roberts (18) suggested that  the H ion was similar to the ammonium ion in i t s effects on absorption of other bases.  Under conditions of high H concent-  ration Hoagland and Broyer (30) found evidence of Ca loss from the plant.  They found that the accumulation of anions and  cations was substantially increased when the pH was raised.  With  increasing dosage of nitrate, leaf Ca of the blueberries increased. This was true for K and :Mg too, thus supporting the work of Sideris and Young (43) on pineapple. Increased rates of ammonium, however, did not lower the absorption of any of the bases significantly. Merril (39) found leaf scorch appearing at low pH values, which was symptomatic of toxicity and this condition was corrected by raising the pH above 4.0 by the addition of lime. In the present study symptoms of nitrate toxicity appeared very early and the severity of the symptoms in the fertilized treatments appeared to be high at higher pH levels (Figure 19).  In most of the previous investi-  gations , i t has been attempted to specify an optimum pH range from pH 4.5 to pH 4.8.  Bailey (5) even suggested that liming a s o i l at  pH 4,6 was harmful. The results of the investigations carried out here are not in accord with the above findings.  No harmful effects  74  J 3.4  AZ  53-  NO3- 40 IBS W T » ACKf LOW VfiKTER TfeLE i  Figure 19A Leaf symptoms of Nitrate toxicity on Bluecrop Blueberry grown i n pot-culture at 4 pH levels with a high water table. Note yellowing of leaves at low pH levels and leaf scorch at higher pH levels. Leaf size has no bearing on treatment effects, as leaves shown here are from different position on shoots. t  t  Figure 19B L'eaf symptoms of Nitrate toxicity on Bluecrop blueberry grown i n pot-culture at 4 pH levels under free-drained conditions. Leaf scorch symptoms were more severe at higher pH levels. Leaf size has no bearing on treatment effects, as leaves shown here are from different positions on shoots.  75  NH.-N  NO3-N  Figure 19C A comparison of the effect of 60 lbs. N per acre applied asawmomui^nd mWe N to 'Bluecrop' blueberry grown i n peat at pH 5.2. Note severe leaf scorch symptoms i n the nitrate treatment (right).Ammonium treated plant (left) shows only mild symptoms.  NH^-N NO3-N Figure 19D The comparison of the effect of 60 lbs. N per acre applied as ammonium andnftwHtN to 'Bluecrop' blueberry grown in peat at pH 6.2. Toxicity symptois were observed on nitrate treatment (right). Ammonium treated plants show no symptoms of toxicity at this stage of growth although symptoms were observed immediately after treatment.  - 76 due to pH alone were observed in the peat even when the pH was raised to 6.0. These observations seem to be in agreement with the results of Stene (44), Boiler (8) and Cain (11), who a l l claimed that blueberries could be successfully grown even at pH 6.5 provided there was a proper nutrient balance.  The ammonium  form of N is believed to be essential at high pH levels, and the nitrate in acid soils is usually converted to the ammonium form by s o i l denitrifiers.  Cain and Eck (14) have suggested that this  conversion of nitrate to ammonium could be the reason for success of the blueberry in acid soils. Soil pH and leaf tissue pH seem to be closely interrelated (13).  High s o i l pH values tend to promote the uptake of  substantially large amounts of basic cations resulting in a higher leaf pH. Cain (13) found that this condition resulted in chlorosis, which could be prevented by increasing s o i l acidity or creating better conditions in the nutrient environment. Although pH was adjusted in the present .experiment no chlorosis symptoms were noted. The toxicity of higher levels of nitrate was, however, more prevalent at higher pH values in both high and low water table treatments.  - 77 V.  SUMMARY  The effect of waterlogging on the growth and mineral composition of the highbush blueberry was investigated using one year old plants of the variety "Bluecrop" grown under greenhouse conditions. Under waterlogging, growth retardation was observed in a l l the parameters studied: namely shoot length, shoot number, leaf number and leaf area.  Significant inter-  actions were obtained between water table and pH on the growth factors studied. Shoot length was affected by the rate and source of N under the two water regimes.  At lower concentra-  tions of both nitrate and ammonium forms of N, the adverse effect of waterlogging manifested itself in the form of shorter and more compact plants. This was evident i n the check as well as the treatment receiving 20 lbs. per acre of N.  Severe yellowing and  premature leaf aging was also characteristic of the waterlogged treatments.  More obvious than any other factor was the effect  of a high water table on leaf nutrient composition.  Leaf N, P,  K, Ca, Mg, and Fe levels in the leaves were a l l significantly lower than in plants maintained under free-drained conditions. A greater disparity was seen between the two systems of water management in October.  The overall reduction of foliar nutrients with  the presence of a high water table in the highbush blueberry, suggests the need for aeration. The deficiency values suggested in Table 1 and the obtained levels (Table 4), show that the high water  - 78 table treatments were deficient in leaf N, K, and Ca.  Of these,  the most pronounced deficiency was that of N, which hardly increased until October when plants were waterlogged. Many workers have stressed the importance of s o i l pH for blueberries. and pH 4.8.  The c r i t i c a l range was presumed to be between pH 4.2 Four pH levels were manipulated by the addition of lime  giving pH levels of 3.4, 4.3, 5.2, and 6.0. growth were significant.  With increase in pH from 3.4 the amount of  shoot growth appeared to increase. obtained at pH 4.3.  Some effects of pH on  A higher leaf number was also  Leaf number and leaf area were, however, i n -  fluenced both by pH and water table as cited earlier.  Except for  small differences in leaf N in October and leaf K in August arid October, levels of no element appeared to be affected by pH. Of the two forms of nitrogen used in the experiment, i t was clearly demonstrated that the ammonium form of N was taken up in larger amounts, giving significantly higher leaf N values. Although leaf N levels increased with increasing doses of N, shoot growth and general performance of the highbush blueberry seemed better with lower doses (20 and 40 lbs.) of N f e r t i l i z e r , especially with the ammonium form of N, cation are v i t a l .  In the nitrate form, s o i l pH and rate of appliIf the level exceeds 20 lbs. N per acre, nitrate  toxicity may result in severe leaf-burn and perhaps ultimate death of the plant.  These symptoms were more severe at the higher pH  levels than when the s o i l was maintained at pH 3.4.  The unfertilized  plants exhibited more multiple deficiency symptoms when the peat was  - 79 not limed, indicating the need for supplementing plant nutrient needs when highbush blueberries are grovm in raw peat. The success of blueberry cultivation depends to a large extent on specific climatic and edaphic factors. From these studies and from previous investigations made elsewhere, i t is apparent that the plant is peculiar in its nutrient requirements. As a member of the Heath family, Ericaceae, the highbush blueberry is distinctive as a group, showing preference for elements that are in preponderance in nutrient environments essentially acid in nature.  - 80 VI.  ACKNOWLEDGMENTS  I wish to thank Dr. G.W. Eaton, Associate Professor, Division of Plant Science, University of British Columbia, under whose supervision this project was undertaken, for his technical advice during the research, and for his guidance in the preparation of this thesis. Grateful acknowledgment is also extended to the other members of my thesis committee:  Dr. Dr. Dr. Dr.  V.C. Brink, A.J. Renney, CA. Hornby, D.P. Ormrod,  Division Division Division Division  of Plant of Plant of Plant of Plant  Science Science Science Science  My special thanks to the Blueboy Blueberry Company for providing the plants used in this project. The research was supported in part by the C.D.A. Operating Grant No. 48 awarded to Dr. G.W. Eaton, the British Columbia Blueberry Co-op Association, the Colombo Plan and a University of British Columbia Graduate Fellowship.  - 81 VII.  LITERATURE CITED  Ackley, W.B., P.C. Crandall and T.S. Russell 1958. Use of linear measurements in estimating leaf areas. Proc. Amer. Soc. Hort. Sci. 72: 326-330. A.O.A.C.  Association of Official Agricultural Chemists. Atomic absorption spectrophotometry. Official methods of analysis. 10th Ed. (1965) 23-24.  Addoms, R.M,, and F.C. Mounce 1931. Notes on the nutrient requirements and the history of the Cranberry (Vaccinium macrocarpum) with special reference to mycorrhiza. PI. Phys, 6: 653-68. Bailey, J.S. 1936. A chlorosis of cultivated Blueberries. Proc. Amer. Soc. Hort. Sci. 34: 395-6. »  1941. The effect of lime applications on the growth of cultivated Blueberry plants. Proc. Amer. Soc. Hort. Sci. 38: 465-7. > and J.N. Everson 1937. Further observations on a chlorosis of the cultivated Blueberry. Proc. Amer. Soc. Hort. Sci. 35: 495-6. » C.T. Smith and R.T. Weatherby 1949. The nutritional status of the cultivated Blueberry as revealed by leaf analysis. Proc. Amer. Soc. Hort. Sci. 54: 205-8. Boiler, CA. 1951. Growing Blueberries in Oregon. Oregon Agr. Exp. Sta. Bui. 499. Bould, C. 1966. Leaf Analysis of deciduous fruits. In Fruit Nutrition. Second edition. Edited by N.F. Childers, Somerset Press, Somerville, New Jersey, p. 651-684. Buttery, B.R., W.T. Williams and J.M. Lambert 1965. Competition between Glyceria maxima and " Phragmites communis in the region of Surlingham Broad. 11. The fen gradient J. Ecol. 53: 183-95.  - 82 11.  Gain, J.C. 1952. A comparison of ammonium and nitrate nitrogen for Blueberries. Proc. Amer. Soc. Hort. Sci. 59: 161-66.  12.  , and G.J. Galletta 1954. Blueberry and Cranberry. In Fruit Nutrition, edited by N.F. Childers, Somerset Press, Somerville, New Jersey, p. 121-152.  13.  , Blueberry leaf chlorosis in relation to 1954. leaf pH and mineral composition. Proc. Amer. Soc. Hort. Sci. 64: 61-70.  14.  , and P. Eck  1966. Blueberry and Cranberry. In Fruit Nutrition. Second edition. Edited by N.F. Childers, Somerset Press, Somerville, New Jersey, p. 101-129. 15.  Chandler, F.B. 1938. The effect of lime on the lowbush Blueberry. Proc. Amer. Soc. Hort. Sci. 36: 477.  16.  Chapman, H.D.,and P.F. Pratt 1961. Methods of analysis for soils, plants and waters. Div. Agr. Sciences, University California.  17.  Clark, H.E, 1941, Growth and composition of the strawberry as affected by source of nitrogen and pH value of the nutrient medium. New Jersey Agr. Exp. Sta. Bui. 691.  18.  Colgrove, M.S., and A.N. Roberts, 1956. Growth of the Azalea as influenced by ammonium and nitrate nitrogen. roc. Amer, Soc. Hort. Sci. 68: 522-36.  19.  Coville,^ F.W. 1910. Experiments in Blueberry culture. Dept. Agr. Bui. 193.  20.  Darrow, G.M., and J.N. Moore 1962. Blueberry growing. U.S.D.A, Farmers Bui. No. 1951.  U.S.  - 83 21.  De Long, W.A. 1965. Nitrogen nutrition of woody crops. Unpublished mimeographed., ^.review. Can.. '•!<•. Dept. Agr.'. jReiv-Sta.^ 3-r,-' t^',)., Kentville, Nova Scotia.  22.  Dickman, S.R. and R.H. Bray Colorimetric determination of phosphate. Ind. Eng. Chem., Anal. Ed. 12: 665-68.  23.  Doelhert, C.A. 1937. Blueberry Tillage Problems and a New Harrow. N.J. Agri. Exp. Sta. Bull. 625.  24.  >  1944.  25.  Fertilizing commercial Blueberry fields. in New Jersey. New Jersey Agr. Exp. Sta. Circ. 483.  , and J.W. Shive 1936. Nutrition of the Blueberry (Vaccinium corymbosum) in sand culture. Soil Sci. 41: 341-50.  26.  Dumenil, L. 1961. Nitrogen and phosphorus composition of corn leaves and corn yields in relation to c r i t i c a l levels and nutrient balance. Proc. Soil Sci. Soc. Am. 25: 295-98.  27.  Elwell, W.T., and J.A.F. Gidley 1962. Atomic absorption spectrophotometry. MacMillan Co. New York. p. 25-40. :  28.  Gore, A.J.P., and C. Urquhart 1966. The effects of waterlogging on the growth of Mplinia caerulia and Eriophorum vaginatum. J . Ecol. 54: 617-633.  29.  Hall, I.V., L.E. Alders, and L.R. Townsend 1964. The effect of s o i l pH on the mineral composition and growth of the lowbush Blueberry. Ca. J. Plant Sci. 44: 433-38.  30.  Hoagland, D.R. and T.C, Brpyer 1940. Hydrogen ion effects and the accumulation of salts by barley roots as influenced by metabolism. Amer. J. Botany. 27: 173-185.  31.  Holley, K.T., T.A. Picket, and T.C. Dulen 1931. A study of ammonium and nitrate nitrogen for cotton. I. Influence on absorption of other elements. Georgia Agr. Exp. Sta. Bui, 169.  - 84 32.  I l j i n , W.S. 1951. Metabolism of plants affected with lime induced chlorosis. II. Organic acids and carbohydrates. Plant and Soil. 3: 339-351.  33.  Jain, T*C, and D.K. Mi.sra, 1966. Leaf area estimation by linear measurements in Ricinus communis. Nature. 212 (No. 5063) : 739-40.  34.  Johnston, S. 1942. The influence of various soils on the growth and productivity of the highbush Blueberry. Mich, Quart. Bui. 24: 307-310.  35.  Render, W.J., and W.T. Brightwell 1966. Environmental Relationships. In Blueberry Culture. Edited by Eck, P., and N.F. Childers. Rutgers University Press, New Brunswick, New Jersey, p.75-93.  36.  Kramer, A., and A.L. Schrader <• 1945. Significance of the pH of Blueberry leaves. Pi. Phys. 20: 30-36.  37.  Labanauskas, C.K,, J. Letey, L.J. Klotz and L.H. Stolzy 1966. Influence of irrigation and s o i l oxygen on the nutrient content of citrus seedlings. Calif. Agr. 20 (12) 13.  38.  L i , Jerome C.R. 1965. Statistical inference I. Edwards Brothers Inc. Ann Arbor, Michigan, p. 270-73.  39.  Merril, T.A. 1939. Acid tolerance of the highbush Blueberry. Mich. Agr. Exp. Sta. Quart. Bui. 22: 112-116.  40.  Oertli, J.J. 1963. The effect of N and pH on the growth of Blueberry plants. Agron. J. 55: 305-307.  41.  Rayner, M.C. 1925. Nutrition of mycorrhiza plants. Exp. Biol. II. p. 265.  42.  Smith, P.F. 1966. Leaf analysis of citrus. In Fruit Nutrition. Second Edition. Edited by N.F. Childers, Somerset Press, Somerville, New Jersey, p.208-228.  Brit. Jour.  - 85 43.  Sideris, CP., and H.Y. Young 1946. Effects of nitrogen on growth and ash constituents of Ananas comosus (L). Merr. Pi. Phys. 21: 247-270.  44.  Stene, A.E. 1939. Some observations on Blueberry nutrition based on greenhouse culture. Proc. Amer. Soc. Hort. Sci. 37: 620-622.  45.  Willis, L.G., and J.O. Carrero 1923. Influence of some nitrogenous fertilizers on the development of chlorosis in rice. J. Agr. Res. 24: 620-640.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items