Jacobs Journal of Agriculture

Compliance of Finnish Male CHD and Total Mortality with Soil Fertilization in 1957- 1990

*Toysa Timo
Department Of Agriculture, Pohjolank 15, 74100 Iisalmi, Finland

*Corresponding Author:
Toysa Timo
Department Of Agriculture, Pohjolank 15, 74100 Iisalmi, Finland
Email:timo.toysa@fimnet.fi

Published on: 2018-10-15

Abstract

Magnesium (Mg) tissue content has been associated with vascular pathology and risks of coronary heart disease (CHD). Associations between fertilization, morbidity and mortality have been studied in veterinary, but less in human medicine. In this study we assessed changes in male (M) CHD and total (TOT) (mortality) and their regressions by fertilization parameters: calcium (Ca), Mg, potassium (K), phosphorus (P), nitrogen (N) and carbonate (CO3) (direct or functions of these parameters) in 1957-1990, when medical treatments were less effective than today. Fertilization parameters are given as equivalents/ha. The aim of this study is to assess whether mutual ratios of mineral elements in fertilization of Finnish agricultural soils were associated with male TOT and CHD.
Results
(Regressions by) [Mg/(Ca+Mg+K)] and (combined regressions by) [Ca;Mg;K], [CO3;Mg;K] and [N;P;K] explained highly significantly (p < 0.001) and remarkably variation in TOT (94-98 %) and in CHD (57 - 81 %). In the represented combined regressions coefficients by mineral elements promoting Mg uptake (Mg and N) were negative, by mineral elements reducing Mg uptake (Ca, CO3, K and P) generally positive.
Conclusion
The mutual variation in the amounts of Ca, CO3, Mg, K, P and N fertilizers of Finnish agricultural soils in 1957-1990 explained significantly TOT and CHD mortality. Effects of Mg on 300 enzymes could explain its primary effect on TOT and secondary effect on CHD. These associations could be mediated through Mg variation in basic food.

Keywords

Total Mortality; CHD; Calcium; Carbonate; Fertilizers; Magnesium; Potassium; Nitrogen; Phosphorus; Agriculture

Introduction

Decreased tissue magnesium in sublingual epithelial cells [1], and femoral muscles [2] have been reported in connection with vascular pathology. Experimental hypomagnesemia can cause vascular damages and myocardial necrosis in different animals [3]. Additional magnesium intakes of animals on atherogenic, hyperlipidemic diets decreased arterial and myocardial lipid deposition without lowering the elevated serum lipids [4]. In contrast, high calcium intakes have been reported to decrease the serum lipids but to raise the arterial lipids [5]. Atherosclerosis has been strongly associated with oxidative stress and disturbed mitochondrial metabolism and function [6]. This is supported by experiments of Manju and Nair (2006) [7]. CHD risk factors [8] cholesterol [9] and total cholesterol: HDL cholesterol ratio (as LDL/HDL cholesterol) [10] can be seen to some extent as indicators of lipid peroxidation, too.Additionally to the effects on CVD risk factors, coronary artery spasm, cardiac arrhythmia, and increased vulnerability to myocardial necrosis following coronary occlusion, may all be dependent on changes in myocardial and vascular smooth muscle electrolyte metabolism that follow from the reduced extracellular Mg [11]. Effects of Mg on 300 enzymes and difficulties in determining the Mg status of humans and animals [12] reduces the access to epidemiologically reliable data concerning Mg variation in human and veterinary tissues. Here we try to assess Mg-changes in human tissues in period 1957-1990 indirectly via fertilization statistics and their compliance with male mortality statistics (CHD and TOT).

Agriculture-Mineral Elements and Interactions

During thousand millions of years mother-earth has balanced by volcanic eruptions, glacial procedures and sea bottom elevations the mineral element losses of soils caused by erosion. Human beings have fought against mineral losses by recycling the original plant nutrients and different fertilizers more than 2,000 years [13]. The general principle known as the Liebig’s rule: "The availability of the most abundant nutrient in the soil is only as good as the availability of the least abundant nutrient in the soil" [14], does not work in details: In addition to their special functions, cations Ca, Mg, K and Na can in certain limits replace each others [15]. The equivalent sum of these cations per dry weight of soil in standard circumstances is approximately constant, and by changing their ratios in soil it is possible to change their ratios in plants [15]. Liming (mainly CaCO3) can reduce the availability of (neutral ammonium acetate extractable) Mg and change a part of extractable Mg to non-extractable Jokinen (1981) [16]. CO3 can increase leaching of cations, including Mg [17]. Phosphates can precipitate Mg with ammonium to a very slightly soluble MgNH4 PO4 ?6H2 O [18], and reduce its availability to plants and animals. Nitrates can increase the Mg content of plants [19]. 

Cattle Breeding

Grass tetany risk is known to be increased if equivalent ratio of K/(Ca+Mg+K) in fodder exceeds 2.2 [20]. High amounts of K and N fertilization is known to decrease the serum Mg contents in cow [21].

Materials and Methods

Supplementation data of N, P and K mineral fertilizers per ha 1951-1960 are from Sillanpää (1978) [22]. Consumption of N, P and K mineral fertilizers in 1961-1999 are from FAO [23]. The area of cultivated arable land in 1951-1960 is from Official Statistics of Finland [24] and arable land in 1961-1999 from FAO [25]. The amounts of Ca, Mg and CO3 in liming agents (carbonates) in 1951-1999 are from Nordkalk [26]. In the study period (1957-1990) mean supply of Ca was 86 and Mg 13 kg/ha/a from carbonates. During the same period Ca from phosphates, approximated according to the formula of superphosphate by multiplying the P-values by 0.65, was ca 17 kg/ha/a. Ca(NH4 )(NO3 ) and in Ca-nitrate gave 1.6 kg Ca/ha on the average (FAOSTAT) [23,25]. The same value as in 1961 (0.7 kg/ha) [23,25] has been approximated for 1957-1960. Mg supplementation from NPK-fertilizers for 1957 (1.7 kg/ha) is from Heinonen (1956) [27], for 1970-1980 from Jokinen [28], for 1981-1999 from Statistics Finland [29, 30] and for 1957- 1970 linearly interpolated (1.7 – 2.3 kg/ha/a). Mg deposition 0.8 kg/a [28], was included in mineral Mg fertilizers, too. (Table 1).

Table 1. Summary of mean and range of Ca, Mg, K, P and N mineral fertilization during 1957-1990 in Finland

Approximation of Ca and Mg are more freely approximated for period 1951-1956. Death-rates of TOT and CHD for 1951-1968

Table 2. Male age adjusted total (TOT) and CHD mortality of men 35-64 y old and annual fertilization of calcium (Ca), magnesium (Mg), potassium (K), phosphorus (P), nitrogen (N) and carbonates (CO3) as equivalents/ha

are from Valkonen and Martikainen [31] and for 1969-1999 from Statistics Finland [32]. Data (Table 2) are represented from 1951-1999, but the calculations concern only the period 1957-1990. In Figure 1 are represented relative values of nonCHD (= TOT - CHD), too.

Visual Assessment of Mortality Mortality

statistics are presented as proportional changes as per cents to the year 1957 [Δ.(.i;.t0 ) %], when (CHD/TOT) ratio got its lowest value of the 1950’s (Figure 1). TOT had its lowest value of the 1950's in 1959, reached its maximum in 1969, declined below the level of the 1950's in 1972 and had a stagnation period since 1983. The CHD reached its maximum in 1967. The absolute epidemic of CHD mortality ended in 1983, when the CHD mortality declined below the level in 1957. The end of the proportional CHD epidemic (tω) occurred within 1993 when (CHD/TOT) reached (CHD/TOT).1957.

Figure 1. Relative changes in male CHD total and non-CHD mortality in Finland 1951-1999

Figure 1 shows age adjusted male CHD, TOT and non-CHD death-rates (1/100,000) of 35-64 year old men relative to their values in 1957, as well as the exponential line("Trend") connecting their crossing points in 1957 and tω

Visual Assessment of Fertilization Statistics

Because of simplicity liming agents (and their components, including CO3) here and later are included below the label of "fertilizers". In figures some years, mostly 1967, 1969 and 1983 are marked with gray lines for clarifying the text. In Figure 2 are for comparison consumption of Ca, Mg and CO3 mineral fertilizers (fm). All of them reached their maximum in 1983 and were declining after that. The amount of Ca fertilizers (from several sources), was higher than carbonates until 1975, after that CO3 exceeded it via increased use of dolomite [31] and decrease of P fertilizers. (Figure 2, Figure 4).

Figure 2. Supplementation of Ca, Mg and CO3 on Finnish agricultural soils in 1951-1999

Figure 3 shows changes in consumption of N, P and K fertilizers: In 1951-1960 phosphorus presented the highest proportion, since 1961 consumption of N exceeded it and increased most rapidly. In 1961-1998 the second highest proportion of NPK fertilizers came from phosphates. P reached its maximum in 1974, K 1983 and N 1989. N, P and K declined after 1989. The variation in components of NPK fertilizers was similar.

Figure 3. Supplementation of N, P and K on Finnish agricultural soils in 1951-1999

Figure 4 shows ratios of other main fertilizers to Mg and ratio of P to N. Ca/Mg and CO3/Mg had their tops in 1964 (before CHD maximum). K/Mg and P/Mg reached their maximum in 1969 and declined below the level of 1957 until 1975 and continued declining until 1984. They increased slightly between 1984- 1987 and reached the level of 1984 until 1990. In 1969 K/ Mg and P/Mg are associated with simultaneous absolute maximum of TOT and simultaneous relative maximum (to trend-line) of non-CHD. In 1983-90 K/Mg and P/Mg stagnation (with Mg decrease), are associated with absolute stagnation in non-CHD and relative stagnation of TOT. Changes in N/Mg were similar to K/Mg and P/Mg, but delayed. P/N declined nearly exponentially (linearly on the logarithmic scale)

Figure 4. Changes in quantitative ratios of N, K, P, Ca and CO3 to Mg and P to N in fertilizers relative to their values in 1957 in Finland

Statistical Survey

[Mg/(Ca+Mg+K)] associated negatively and highly significantly (p < 0.001) with TOT and CHD. It "explained" TOT by 95 % and CHD by 57 %. Other associations - [Ca/(Ca+Mg+K)] and [K/(Ca+Mg+K)] - were positive and significant (p < 0.025), but weaker. (Table 3).

Table 3. R squares of male CHD regressions by fertilization parameters with significances and signs of correlation coefficients.

In Table 4 [Ca;Mg;K] and [Ca;Mg;NPK] explained highly significantly (p < 0.001) TOT and CHD. Replacing Ca by CO3 in these regressions increased their explanative strength. Coefficient signs of Ca and CO3 were always positive and by magnesium negative. Coefficient signs of K were the same as by NPK in respective regressions, positive or negative. In regressions by [Ca;Mg;N;P;K] and [CO3;Mg;N;P;K] coefficient signs of Mg and N were negative, by Ca and CO3 positive and by K and P variable. Regressions by [N;P;K] explained highly significantly TOT and CHD, coefficient signs of P were positive and by N and K negative.

In the represented regressions (Table 4) coefficient signs of Mg and N were always negative, by Ca and CO3 positive and by K and P variable (Table 5).

Discussion

Visually absolute TOT epidemic (1959-1972) occurred totally and CHD epidemic (1957-1983) (Figure 1) by its most part within the range of relative excess of K, P and N to Mg fertilization. Remarkable is the stagnation in decline of TOT and non-CHD mortality in 1983-90 associated with decrease of Mg fertilization and stagnation in K/Mg, P/Mg and N/Mg fertilization ratios. During period 1957-1990 CHD was increasing for 10 and TOT for 12 years, CHD declined for 23 years and TOT for 21 years.When fertilization was generally increasing during this period negative coefficient signs in regressions by single parameters are expected. From six fertilization parameters is possible to form 63 different combinations and more their mathematical functions (sums, ratios), but biological importance of all such combinations or functions are not known. [Mg/(Ca+Mg+K)] and multiple regression by [Ca;Mg;K] explained 95 % of TOT and 78 - 57 % of CHD (p < 0.001 for all of them). Coefficient signs of Mg and [Mg/(Ca+Mg+K)] were negative and by Ca and K positive. The mineral elements given in fertilizers are additive to the soil reserves and recycled nutrients. Prerequisites for Mg effects were a delay in Mg supplementation, delay in reaching the previous soil Mg level by fertilization and a change in competition with other mineral elements (Figure 2-4).

Potassium

Interactions of Ca, Mg and K are widely studied in soil science at least since the 1930’s [15]. In period 1957-1990 [K/ (Ca+Mg+K)] associated significantly positively with TOT (p = 0.002) and highly significantly positively (p < 0.001) with CHD (Table 2). These “effects” can possibly be explained by its competitive effects on plant Mg-uptake [15]. In combined regressions coefficient of K was variable, possibly because of selected period. In human medicine the low serum K is known to be treated by Mg [33], but the Mg decrease caused by excess K [20,21] seems not be known in PubMed. Because K fertilization of pasture is known to decrease the serum Mg content in cow [20], we can suppose that such response is possible in humans especially treated with diuretics and high K substitution.

Calcium and Carbonate

Besides of competition [15] effects of excessive Ca (strongly associated with CO3 ) could be explained by the reduction of Mg availability from soil and change of Mg given in fertilizers to non-extractable form [16], by leaching [17] and by the antagonistic effect on Mg liberation from soil reserves [34]. Ca/(Ca+Mg+K) associated highly significantly positively with TOT (p < 0.001) and less significantly CHD (p = 0.022). Replacing Ca by CO3 in CHD regression by [Ca;Mg;K] increased its explanative strength (Table 3). Excessive Ca intake has been reported to increase risk of cardiovascular events [35], in concordance with Vitale et al [5]. But changes of Ca proportion in plants and foodstuffs could be of limited importance because generally is known that Ca intake by humans is mainly depending on foodstuff selection (e.g. inclusion or exclusion of milk products). Carbonate effects could be mediated via silicon (Si) and P, too (in next chapters). The possible harmful effects of high CO3 rates have been largely overcome by the change of liming agents to more Mg containing dolomites [28]. Exact distribution of phosphor fertilizers subgroups in the 1950's and 1960's is difficult to determine, maybe possible through the data in National Archives, but not sure [36]. So Ca supplementation via phosphates has been given value, determined by superphosphates, which obviously underestimates, possibly mostly in the 1950's, but does not remarkably change its "positive" association with mortality. 

Phosphor

Combined regressions by [N;P;K] explained 94 % of TOT and 73 % of CHD. Coefficient signs of P were positive and by others negative. Veterinary surgeon Pekka Nuoranne supposed that in Finland the high P:Ca ratio in human diet could be more inconvenient than high Ca intake and danger of excess allowance of Ca from normal food were non-existent [37]. In the 1970’s Finnish diet Ca:P ratio was 1500:2000 (mg/d) [38]. 2012 this ratio was ca 1100/1500 [39]. Intake of phosphor was more than two-fold to RDA [39]. P-excess to Ca and Mg in plasma are reported to be associated with tetany [40], and so obviously with increased risk of fatal CVD. Harmful associations of P fertilizers with death-rates could be explained by Mg precipitation [18]. Sulphur, included in superphosphate, could decrease Se-uptake and promote Se deficiency [41], so especially in the 1960’s and 1970’s, when the P-fertilization was on its highest level. CO3 dressings are recommended for increasing the soluble P content and to keep the supplemented P available to plants [42]. So we can suggest that the strong peaks of CO3 supplementation (Figure 2) could have worked as additional P impulses, too.

Nitrogen

In TOT and CHD regressions by [N;P;K] coefficient signs of N were negative (Table 3) and could be mainly be explained by its positive effect on Mg-uptake by plants. In 1957-1990 the mineral N-fertilization was small (mean 63 and maximum 108 kg/ha), remarkably below of the level of normal for grass fertilization in Finland in the 1970's, which was about 150 - 300 kg/ha [43] and tested in the Netherland in the 1950's (210 - 230 kg/ha) [21], thus N-fertilization rates on non-grass plants were remarkably lower. The antagonistic effects of high N-fertilizer rates on S-Mg [21] of cows are possibly mediated via their complicated digestive system and are supposedly different as the N effects in humans. Possibly (not uncommon?) insufficient N-fertilization could not counteract the negative effects of K on Mg uptake [19]. The antagonistic effect of NH4+ on Mg uptake is known [44], but the total amount of urea N and as NH4+ ions was in 1961-1990 small, mean 3.5 kg N/ha (ca 5 % of mineral N-fertilization), with "sharp" maximum in 1973 (7 kg N/ha) [23,25]. This maximum did not associate with any visible effect on CHD, but in 1974 we see an elevation in TOT (Figure 1). One explanation of N and P associations with death-rates can be the historical change in P/N ratio (Figure 4). Additionally statistical differentiation of N, P and K "effects" is difficult because they are so similarly time-related.

Silicon and Selenium

Silicon is not included in routine fertilization in Finland as in western countries [45], and increasing mineral/recycled fertilization (fm/rcl) ratio in 1957-1984 obviously indicated reduction of plant available silicon (Figure 5). So it is not a surprise that Finnish grain contained less than 0.1 g Si/kg [46], which is remarkably less than reported in Australia (0.5-5 g/ kg) [47]. The obvious Si decline increased need of selenium supplementation [48,49]. Jones and Handreck (1965) reported that plant Si content associated inversely with soil pH [50], which supports the role of acidity in Mg availability, because Mg and Si often are liberated from the same sources [36]. So CO3 fertilization could have been caused fluctuation in Si plant uptake and Si intake by livestock and humans.

Recycling etc

Recycled mineral elements (rcl) have been discarded from the statistical survey because of discrepancy concerning their estimated amounts. In recycling occurred several timerelated and area-related changes in 1957-1990. N-fertilizing via biological N-synthesis was important in the 1950’s, but it became unimportant in the 1960’s via disappearance of clover: in 1966-1967 clover proportion in hay crops was reduced to 7-8 % [51]. This N as well as soil own (exchangeable and acid soluble) mineral elements are outside of this survey. The amount of recycled NPK fertilizers has been suggested to have been slightly decreasing in 1957-1975 [22]. On the other hand the amount of Mg.rcl has been estimated to be increased from 5 kg/ha in 1956 [52] to 7 kg/ha in 1981 [16]. It is possible to suppose that the K.rcl and Mg.rcl were rather similarly timerelated, why in Figure 5 we have benefited fixed value 22 kg/ha for K.rcl, slightly modified from [22]. Mg.rcl was given value 6.5 kg/ha, as a compromise between [16,52]. (Mg deposition was included here in mineral Mg fertilizers, but could be included in Mg.rcl, possibly to losses too, if supposed that their main sources are the agricultural soils). For Ca.rcl was approximated value 15 kg/ha from several Finnish sources. For N is given value 19 kg/ha from Sillanpää as it was in 1962 [22]. (Table 6).

Table 6. Recycled plant nutrients kg/ha/a (approximated)

Figure 5 shows changes in ratios of mineral elements (fm) in mineral fertilizers to their respective amounts in recycled fertilizers (fm/rcl). These parameters have similar variation as pure mineral fertilizers, because they have been given constant values. Ratio Mg.(fm/rcl) was 1957-1969 about one and exceeded permanently this value first ca 1970. We see delay in Mg.(fm/rcl) to K.(fm/rcl) before 1975, and to P.(fm/ rcl) before 1982.

Figure 5. Development of commercial/recycled ratio in Finnish plant nutrients 1957-1998

CHD regression by mere year gave R square 0.350 (p < 0.001), i.e. year "explained" male CHD by 35 % (Figure 6). TOT regression by year "explained" TOT by 85 % (Figure 7).

Figure 6. Male CHD and its regression by [year] from 1957-1990 with SD's and trendline

Figures 6 and 7 suggest on moderately strong pure associations between time (year) and death-rates. So all other timerelated factors are potential participants in explaining the development of male TOT and CHD death rates. Remarkable was the stagnation in TOT (as well as in non-CHD) mortality between 1983-1990, which obviously has not earlier got any explanations.

Figure 7. Male TOT and its regression by [year] from 1957-1990 with SD's and trend-line

Conclusion

The mutual variation in the amounts of Ca, CO3, Mg, K, P and N fertilizers of Finnish agricultural soils in 1957-1990 explained significantly TOT and CHD mortality. Effects of Mg on 300 enzymes could explain its primary effect on TOT and secondary effect on CHD. These association could be mediated through Mg variation in basic food.

Acknowledgement

We are grateful to veterinary surgeon Seppo Haaranen for several discussions, as well as Mikko Lauronen, Purchasing Manager of Nordkalk Corporation, for his great work in analyzing the data concerning changes in Mg and Ca contents of liming agents in Finland.

Examples of Regressions

Regression by [N;P;K] explained 83 % of male CHD variation in 1961-1990 (Figure 8). Coefficients of N and K were negative and by P positive. (The data concerning fertilization and arable land are freely accessible from FAOSTAT [23,25]).Figure 8 shows CHD regression by[N;P;K]. Its top, at 1968, is very close the real CHD maximum.

Figure 8. Male CHD and its regression by [N;P;K] fertilization, with SDs and trend-line in 1961-1990

Figure 8 shows CHD regression by[N;P;K]. Its top, at 1968, is very close the real CHD maximum.

Other Examples of CHD and TOT Regression

Figure 9. Male TOT and its regression by [Ca;Mg;K] from 1957-1990 with SD's and trend-line

Figure 9 shows TOT regression by [Ca;Mg;K] in 1957-1990. Remarkable is the stagnation since 1983 in TOT, as well as in its regression, one year before by the relative single mineral element ratios to Mg (Figure 4).

Figure 10. Male CHD and its regression by [CO3;Mg;N;P;K] from 1957-1990 with SD's and trend-line

Figure 10 shows CHD and its regression by [CO3;Mg;N;P;K] from 1957-1990 with SD's and trend-line. The regression-line is less coherent than TOT regression by the same parameters (Figure 11).

Table 9. Parameters of CHD regression in 1957-1990 by annual consumption of CO3, Mg,N, P and K fertilizers (Figure 10)

Figure 11. Male TOT and its regression by [CO3;Mg;N;P;K] from 1957-1990 with SD's and trend-line

Table 10. Parameters of TOT regression in 1957-1990 by annual consumption of CO3, Mg, N, P and K fertilizers (Figure 11)

References

1., Shechter M et al. Oral magnesium therapy improves endothelial function in patients with coronary artery disease. Circulation. 2000, 102(19): 2353-2358.

2. Jeppesen BB. Magnesium status in patients with acute myocardial infarction: a pilot study. Magnesium. 1986, 5(2): 95-100.

3. Seelig MS. Magnesium deficiency in the pathogenesis of disease. 1980.

4. Vitale J, White PL, Nakamura M, Hegsted DM, Zamcheck N et al. Interrelationships between experimental hypercholesteremia, magnesium requirement, and experimental atherosclerosis. J Lab Clin Med 53:433-441, 1957. (in Seelig MS. Magnesium deficiency in the pathogenesis of disease, part ii magnesium deficiency in the pathogenesis of cardiovascular diseases. Chapter 6.2.1. Experimental Arteriosclerosis of Magnesium Deficiency, Chapter 6.2.1.3. Arterial Damage of Magnesium Deficiency Intensified by High Fat Intakes. pdf-page 145-503.

5. Vitale JJ, White PL, Nakamura M, Hegsted DM. Effect of feeding an atherogenic diet on magnesium requirement. Fed Proc 16:400-401. 1957. (in Seelig MS. Magnesium deficiency in the pathogenesis of cardiovascular diseases Part II, in Chapter 6.2.1. Experimental Arteriosclerosis of Magnesium Deficiency, Chapter 6.2.1.3. Arterial Damage of Magnesium Deficiency Intensified by High Fat Intakes. 145-503.

6. Finsterer J. Is atherosclerosis a mitochondrial disorder? Vasa. 2007, 36(4): 229-240.

7. Manju L, Nair RR. Magnesium deficiency augments myocardial response to reactive oxygen species. Can J Physiol Pharmacol. 2006, 84(6): 617-624.

8. Giannuzzi P, Wood DA. Eur J Cardiovasc Prev & Reh, 14, Suppl 2, score points on pages. 2007, E12 - E14.

9. Memon RA, Grunfeld C, Moser AH, Feingold KR. Tumor necrosis factor mediates the effects of endotoxin on cholesterol and triglyceride metabolism in mice. Endocrinology. 1993, 132(5): 2246-2253.

10. Bourdon E, Loreau N, Lagrost L, Davignon J, Bernier L et al. Differential effects of oxidized LDL on apolipoprotein AI and B synthesis in HepG2 cells. Free Radic Biol Med. .2006, 41(5): 786-796.

11. Bloom S. Effects of magnesium deficiency on the pathogenesis of myocardial infarction. Magnesium. 1986, 5(3- 4): 154-164. 12. Elin RJ. Magnesium: the fifth but forgotten electrolyte. Am J Clin Pathol. 1994, 102(5): 616-622.

13. Broome SW. History of soil fertility. North Carolina State University.

14. Liebig's law of the minimum.

15. Itallie TB van. Cation equilibria in plants in relation to the soil. Soil Sci. 1938, 46(3): 175-186.

16. Jokinen R. Requirements for Magnesium fertilization in Finland. Journal of the Scientific Agricultural Society in Finland. 1981, 53(4): 239-268.

17. Liming (soil).

18. Pauling L. College Chemistry. WH Freeman Company. San Francisco and London, 1964.

19. Jerlström H-G. Studier över möjligheterna att med växtoch jordanalyser beskriva magnesium-situationen i svensk växtodling, 1975.

20. Kemp A. Grass tetany in grazing milking cows. Netherlands. J Agr Sci. 1957, 5: 4-17.

21. Kemp A. Neth J Agr Sci. 1960, 8: 28-304.

22. Sillanpää M. Lannoitus ja maan ravinnevarojen kehitys. Käytännön Maamies. 1978.

23. FAOSTAT, mineral fertilizers 1961-2002.

24. Use of arable land 1951-60: Official Statistics of Finland 1961.

25. FAOSTAT, arable land 1961-2012.

26. Lauronen M. Nordkalk Corporation. Liming agents, total and their proportional variation in Mg and Ca contents 1949- 1999.

27. Heinonen R. Magnesiumin tarpeesta Suomen pelloilla. Agrogeologisia Julkaisuja No 65. Helsinki, 1956.

28. Jokinen R. Requirements for Magnesium fertilization in Finland. Journal of the Scientific Agricultural Society in Finland. 1981, 53(4): 239-268.

29. Sales of fertilizers to farms. Official Statistics of Finland (Suomen Tilastollinen Vuosikirja) 1981-1990.

30. Sales of fertilizers to farms. Official Statistics of Finland (Suomen Tilastollinen Vuosikirja) 1991-2000.

31. Valkonen T, Martikainen P. Development of mortality from ischaemic heart disease in subgroups of the population in Finland. Journal of Social Medicine. 1990, 27: 273-288.

32. Age adjusted total and coronary mortality (I20-I25) of men and women, 35-64 years, Finland, 1969-2005.

33. Mikkola E. Luke, Natural Resources Institute Finland, 2016.

34. Sheehan JP, Seelig MS. Interactions of magnesium and potassium in the pathogenesis of cardiovascular disease. Magnesium. 1984, 3(4-6): 301-314.

35. Boyle JR, Voigt GK, Sawhney BL. Chemical weathering of biotite by organic acids. Soil Sci. 1974, 117(1): 42-45.

36. Bolland MJ, Avenell A, Baron JA, Grey A,MacLennan GS, Gamble GD et al. Effect of calcium supplements on risk of myocardial infarction and cardiovascular events: metaanalysis. BMJ. 2010, 341: c3691.

37. Nuoranne P. Magnesium deficiency [in Finnish “Magnesiumin puutostila”]. Rauma 1998.

38. Varo P, Koivistoinen P. Chapter XII. General Discussion and Nutritional Evaluation. Mineral Element Composition of Finnish Foods: N, K, Ca, Mg, P, S, Fe, Cu, Mn, Zn, Mo, Co, Ni, Cr, F, Se, Si, Rb, Al, B, Br, Hg, As, Cd, Pb and Ash. Acta Agriculturae Scandinavica Supplementum 22. Stockholm 1980.

39. Helldán A, Raulio S, Kosola M, Tapanainen H, Ovaskainen ML et al. Finravinto 2012 -tutkimus - The National FINDIET Survey, 2012.

40. Haaranen S. Observations on hypomagnesemia with spasmodic symptoms on cows, 1968.

41. Liu XW et al. Effects of sulfur on transformation of selenium in soil and uptake of selenium in rape. Article in Chinese. Huan Jing Ke Xue. 2014, 35(9): 3564-3571.

42. Soil pH.

43. Rinne S-L, Sillanpää M, Hiivola S-L, Huokuna E, Effect of nitrogenous fertilization on silicon content of grass. Kehittyvä Maatalous. 1977, 35: 7-10.

44. Claassen ME, Wilcox GE. Comparative Reduction of Calcium and Magnesium Composition of Corn Tissue by NH4-N and K Fertilization. Agronomy Journal. 1973, 66(4): 521-522.

45. Epstein E. The anomaly of silicon in plant biology. Proc Natl Acad Sci U S A. 1994, 91(1): 11-17.

46. Koivistoinen Pekka. Mineral Element Composition of Finnish Foods: N, K, Ca, Mg, P, S, Fe, Cu, Mn, Zn, Mo, Co, Ni, Cr, F, Se, Si, Rb, Al, B, Br, Hg, As, Cd, Pb and Ash. Acta Agriculturae Scandinavica Supplementum 22. Stockholm 1980.

47. Schultz JE, French RJ. Silicon uptake by wheat and its relation to grain yield and water use. Australian Journal of Experimental Agriculture and Animal Husbandry. 1976, 16(78): 123–128.

48. Tang H, Liu Y, Gong X, Zeng G, Zheng B et al. Effects of selenium and silicon on enhancing antioxidative capacity in ramie (Boehmeria nivea (L.) Gaud.) under cadmium stress. Environ Sci Pollut Res Int. 2015, 22(13): 9999-10008.

49. Garg N, Bhandari P. Interactive effects of silicon and arbuscular mycorrhiza in modulating ascorbate-glutathione cycle and antioxidant scavenging capacity in differentially salttolerant Cicer arietinum L. genotypes subjected to long-term salinity. Protoplasma. 2016, 253(5): 1325-1345. 

50. Jones LHP, Handreck KA. Studies of silica in the oat plant. III. Uptake of silica from soils by the plant. Plant and Soil. 1965, 23(1): 79-96.

51. Jäntti A. Runsaan typpilannoituksen hyväksikäyttö laidunsäilörehunurmilla. Karjatalous. 1968, 3: 82-85.

52. Heinonen R. Magnesiumin tarpeesta Suomen pelloilla. Agrogeologisia Julkaisuja No 65. Maatalouskoelaitoksen Maatutkimusosasto. Helsinki, 1956.

Equations

Equation of Trend line (in Figure 1): y = (E.CHD-CHD.1957 )/ CHD.1957 * 100

tω = i + {[(CHD/TOT).i - (CHD/TOT).1957 ]/[ (CHD/TOT).i - (CHD/TOT).(i+1) ], when (CHD/TOT).i ≥ (CHD/TOT).1957 > (CHD/TOT).(i+1); })

k = ln(CHD/CHD.1957 )/(tω - 1957)

(Exact) CHD.ω = {CHD.i + (tω - i)/[(i+1)-i)* [CHD.(i+1) - CHD.i]; when i is the year preceding tω}

(Exact) TOT = {TOT.i + (tω - i)/[(i+1)-i)* [TOT.(i+1) - TOT.i]; when i is the year preceding tω}