Vol. 4:495-501 Determination of barley nitrogen status with chlorophyll meter for high (3-amylase in grains Jari Peltonen and Ari Virtanen Plant Production Inspection Centre, Seed Testing Department, P.O. Box 111, FIN-32201 Loimaa, Finland Jaakko Helenius, Juha Salopelto and Kari Kiltilä Cultor Ltd., Helsinki Esko Eloranta N eson Ltd., Jokioinen The production of P-amylase is of great importance in two-rowed spring barley cv. Kymppi (Hor- deum vulgäre L.) in the Finland, where long-day conditions favour high enzyme activities. Nitrogen (N) fertilization of a crop is the main means of manipulating barley P-amylase activity for industrial purposes. In this study, leaf chlorophyll content determined with a portable chlorophyll meter (Mi- nolta SPAD-502) in the field, was used to predict N availability of cv. Kymppi for p-amylase produc- tion. Critical chlorophyll meter readings (SPAD values) were calculated from data deriving from experiments with various N fertilizer levels using the Cate-Nelson procedure. According to the res- ults of this study it can be stated that the critical SPAD values at pollination (i.e. pollen grains on well-developed stigmatic hairs, GS 52-58) are 37 SPAD units for grain yield and 41 SPAD units for P-amylase activity. The optimum grain yield occurred at 41 SPAD units and optimum P-amylase activity was reached at 45 SPAD units. Determination of leaf chlorophyll content using the chloro- phyll meter led to more appropriate fertilizer application recommendations and subsequently increased P-amylase activity in grains. Grain protein concentration could be an effective diagnostic tool for post-harvest evaluation of grain P-amylase activity in cv. Kymppi. Key words: P-amylase activity, grain protein concentration, grain yield, chlorophyll content, plant tissue N status ntroduction P-amylase is an importantbarley (Hordeum vul- gäre L.) protein, determining the technical qual- ity of starch used for various industrial process- es in Finland (Helenius 1992). Cultivars with high p-amylase activity are often reported to be high in grain protein concentration (Harris and Banasik 1952). However, Helenius (1992) indi- cated that although increases in grain protein concentration led to increased P-amylase activ- ity in grain, p-amylase activity differed among barley varieties: two-rowed Kymppi had - and © Agricultural Science in Finland Manuscript received December 1995 495 AGRICULTURAL SCIENCE IN FINLAND https://www.c-info.fi/en/info/?token=SDp7Ihx9d11PtlNQ.yTojdKaTOMVFnXnbgDBq2A.HEf_J_5oT_i-o48mseP7uAgU7TIvWepOvpfRm5LsLVfIOnrHlGE7ZlSkB6fHHDrNE5wtoGzPnHDnb_7HKjuX_ta_lI0u37kmO88QcEFHUU67BO5cwc_6WN85aVMmyKTVYhGKxu1lAnFyb3edPL2q5Pd5NsxCzdyLVPoUmJqgeoTHsyS7RoxOhuWF2fHuBJL_nNbcgiPPoJNq0PR4ICzaZtKSUT7Qlb40t3_UZ47AzEu-rA8K95h_acbLW-OK1xWeJXclLdK5cvomDPWLVIW7R0uu3Dv52KdlTk9gJm42oDRPYgNEwjHSeaGkeSqcOgcnhsQTfTYVne_FSEZzbXaRaF6sC5TuBBDvyspUd9ZAlT6TUlYLlGxbZ4TNiA still has - the highest (3-amylase activity among barley varieties grown in Finland. Nitrogen (N) fertilization of a crop is the main means of manipulating barley (3-amylase activity for industrial purposes (Hayter and Riggs 1973). Previously, Peltonen et al. (1995) showed that leaf chlorophyll content, determined in the field with a portable Minolta SPAD-502 chloro- phyll meter (developed by the Soil-Plant Analy- ses Development Section of Minolta Camera Company, Ramsey, NJ) 1 , accurately indicated plant N status of small-grain cereals, allowing N fertilizer requirements to be determined ac- curately. Applying N fertilizer on this basis improved production economics and improved the physical input-output ratio during grain yield formation (Peltonen et al. 1995). The objectives of this research were to quanti- fy relationships between yield, grain protein con- centration, (3-amylase activity and SPAD values measured with a Minolta chlorophyll meter. The specific goals were to evaluate the poten- tial for calibrationing the meter to optimize cv. Kymppi N status and predict the need for sup- plemental N fertilization for increasing (3-amyl- ase activity. Finally, the (3-amylase stability of an optimized production system was evaluated using SPAD values. Material and methods Field trials Field trials (Series 1) included five separate ex- periments at Experimental Farms of the Univer- sity of Helsinki (Viikki and Suitia in 1993 and 1994) and at the Experimental Farm of Kemira Ltd, (Kotkaniemi in 1993) with two-rowed spring barley cv. Kymppi. Arow spacing of 12.5 cm and 500 viable seeds nr2 were used. Nitro- gen (NH 4N0,-N) was applied to three 10m 2 rep- licated blocks at 0, 50, 100, 150 and 200 kg N ' Trade and company names are included for the benefit of the readers and donot imply any endorsement or pref- erential treatment of the product by the authors. ha 1 at Viikki and Suitia, and 0, 40, 80, 120 and 160 kg N ha 1 at Kotkaniemi. Nitrogen fertilizer was applied 7 cm deep at sowing. P, K, Ca, S, Mg, Zn and Mn were also applied at sowing ac- cording to Finnish soil test recommendations: phosphorus availability in particular may influ- ence calibration of the SPAD 502 meter (Follet et al. 1992). This program yielded a series of samples with increasing grain yield, grain pro- tein concentration and (3-amylase activity. Stand- ard crop protection procedures (herbicide and fungicide) were implemented. The plots were harvested when grains reached the caryopsis hard stage (GS 92, Zadoks et al. 1974). Yield was determined at 15% grain moisture content. The grain samples from each of the three replicates were combined and mixed thoroughly. Grain protein concentration was de- termined from whole grain meal using the stand- ard method 46-11 of the AACC (1983). (3-amyl- ase activity was determined according to a method described by Helenius (1992). Chlorophyll measurements Chlorophyll meter readings (SPAD values) were taken at the stage of maximum number of flo- rets per ear primordium (FM), coinciding with Zadoks’ GS 37-41, and at pollination (PO), i.e. pollen grains on well-developed stigmatic hairs (GS 52-58). The developmental stage of the in- florescence was determined according to the scale of Waddington et al. (1983), and was con- sidered to be reached when at least 60 % ofplants were at that stage. It has been indicated (Pelto- nen 1992) that these developmental stages are important for timing of supplemental N applica- tions. Thirty randomly selected plants per plot were used for determining the chlorophyll con- tent. Determination of the SPAD value was made from mid-length on the uppermost fully expanded leaf. Statistical analysis of critical readings The critical levels of leaf chlorophyll were es- tablished using Cate-Nelson analysis (Cate and Nelson 1971, Nelson and Andersen 1977). The 496 AGRICULTURAL SCIENCE IN FINLAND Peltonen, J. el aL: Determination ofbarley nitrogen status with chlorophyll meter for high ffamylase in grains Vol. 4: 495-501. Cate-Nelson II model partitioned the data for critical SPAD values into two group in terms of grain yields; N fertilizer responsive and non-re- sponsive. The Cate-Nelson 111 modelpartitioned the data into three groups; N fertilizer respons- ive, transitional and non-responsive. Grainyield was expressed as a percentage yield Y./Ymax x 100 where Y.is the grain yield when N. kg ha' 1 was applied, and Ymax is the maximum grain yield obtained in an experiment. The increase in grain protein concentration (P) and in (3-amylase act- ivity (P) were calculated as^Pi=P,-Poand^p=p j- P O, where P 0 and P 0 is control of grain protein and P-amylase activity, respectively. Actual grain yield, grain protein concentration and p-amyl- ase activity, as functions of the SPAD values, were consistently the best in terms of the coef- ficient of determination (R 2). On-farm trials The accuracy of the critical and optimal SPAD values for optimizing N fertilizer application for maximum P-amylase production was studied using independent barley (cv. Kymppi ) data from 24 on-farm trials (Series 2) in southern Finland in 1995. Nitrogen fertilizer application at sow- ing ranged from 41 to 115 kg N ha 1 . The SPAD values were measured at the pollination (PO) developmental stages as in Series 1. The barley fields were grouped into categories of near op- timum plant N status and excess N, according to the critical SPAD values for Series 1. Results and discussion The temperatures during crop growth in 1993 and 1994 were close to the mean for the period 1961- 1990, but approximately 60 - 80 mm more rain fell prior to heading in 1993 and 1994 than dur- ing 1961-1990. This weather promoted very high grain yields and a very goodresponse to N fertil- izer in Finland. The grain yield, grain protein concentration and P-amylase activity for all ex- periments are presented in Table 1. The max- imum grain yields obtained in Series 1 were 7292 kg ha 1 in 1993 and 6252 kg ha' 1 in 1994. These maximum yield levels were clearly higher than Table 1. Grain yield, grain protein concentration and P-amylase activity as functions of N application rate for spring barley cv. Kymppi grown 1993 and 1994. Location and year Trait N2 N3Nl a N4 N5 Viikki 1993 Grain yield kg ha- 1 3933 6140 6786 7292 7220 Grain protein % 8.8 9.4 10.2 11.5 12.4 P-amylase activity W.K. 80 87 111 149 23587 111 149 235 Suitia 1993 Grain yield kg ha 1 4023 5680 6028 7058 6570 Grain protein % 9.4 9.6 11.9 12.1 13.3 P-amylase activity W.K. 95 115 165 212 267115 165 212 267 Kotkaniemi 1993 Grain yield kg ha 1 2040 4290 4930 5310 5600 Grain protein % 6.9 8.3 11.0 12.9 14.1 P-amylase activity W.K. 77 87 129 200 25387 129 200 253 Viikki 1994 Grain yield kg ha' 1 4355 5449 5951 5799 5598 Grain protein % 9.8 12.1 12.8 13.5 15.8 P-amylase activity W.K. 131 176 229 256 332176 229 256 332 Suitia 1994 Grain yjeld kg ha' 1 3530 5660 6092 6164 6252 Grain protein % 9.4 11.1 13.3 14.3 15.4 P-amylase activity W.K. 102 134 212 231 253134 212 231 253 " N 1 =O,N2 = 50, N 3 = 100, N 4 = 150 and N 5 = 200 kg N ha- 1 at Viikki and Suitia Ml =O, N 2 = 40, N 3 = 80, N 4 = 120 and N 5 = 160 kg N ha- 1 at Kotkaniemi 497 AGRICULTURAL SCIENCE IN FINLAND annual mean yields in Finland, which, accord- ing to Mukula and Rantanen (1989), are about 3000 kg ha' 1 for spring barley. The higher the N fertilizer rate the higher was (3-amylase activity and grain protein concentration. The maximum (3-amylase activity and grain protein concentra- tion values were obtained at Suitia in 1993 and at Viikki in 1994. Critical chlorophyll meter readings Leaf chlorophyll content, determined with the chlorophyll meter, correlated strongly with grain yield at the stage of maximum number of florets per ear primordium (FM), but poorly with grain protein concentrationand (3-amylase activity (not shown in Table 2). In turn, leaf chlorophyll con- tent at the stage of pollen grains on well-de- veloped stigmatic hairs (PO) correlated more strongly with grain yield, grain protein concen- tration and (3-amylase activity. Therefore, the Cate-Nelson analyses were performed only on data taken at this developmental stage. Cate- Nelson 111 model produced significantly higher R 2 values than Cate-Nelson II models. Both models for critical and optimum SPAD units are shown in Table 2. Table 2. Cate-Nelson equations describing grain yield, grain protein concentration and (3-amylase activity in relation to leaf chlorophyll (SPAD) levels, and suggested critical chloro- phyll meter readings for spring barley cv. Kymppi at PO. R 2Equation Grain yield =3576 + 1868 x X, + 733 x X 2 o.B4*** where x, = 0 if <37 SPAD and 1 if>37 SPAD where x 2 = 0 if <4l SPAD and 1 if >4l SPAD Grain proteinconcentration = 9.5 + 4.7 xX, -1.8 xXj o.74*** where x, =0 if <4l SPAD and 1 if >4l SPAD where x2 = oif <45 SPAD and lif >45 SPAD p-arnylase activity = 109 + 143xX,-51 x X 2 o.72*** where x, = 0 if <4l SPAD and 1 if >4l SPAD where x 2 = 0 if <45 SPAD and 1 if >45 SPAD *** Significantat the 0.001 probability level. a PO = stage of pollen grains on well-developed stigmatic hairs, coinciding with Zadoks’ GS 52-58. These results show that the application of sufficient N to spring barley cv. Kymppi, accord- ing to the recommendations based on leafchloro- phyll content (SPAD units) at PO, ranking “low” in N would result in an average response of 3576 kg ha 1 in grain yield as compared with an average response of 1868 kg more grain ha' 1 for the “medium N” class. In class “high", the grain yield would change only by 733 kg ha 1 compared with class “medium". For grain pro- tein concentration and (3-amylase activity the critical SPAD values were 41 units and the op- timum SPAD values 45 units. This indicates that the use of N fertilization in excess of the amount needed for an optimum grain yield (SPAD values exceeding 41 units) will generally result in in- creased grain protein concentration and (3-amyl- ase activity, but not necessarily in grain yield. The likely reason for this is that the photosyn- thetic rate, which is important in determining yield potential, will not increase to the same extent as the nitrate reductase activity of the plant (Lawlor et al. 1987, Joy and Peltonen 1993). In this study, chlorophyll content within the zone of adequacy resulted in 14.2 % of grain protein and 252 W.K. of (3-amylase activity. However, if the SPAD values ranked “high", indicating over optimum N status, there wouldbe a5l W.K. decrease in the (3-amylase activity of grains (Table 2). Effect of optimized N management on (3-amylase production stability using SPAD values Yields were high in 1995 in on-farm trials (Series 2) and generally there were strong N re- sponses. There were also large differences be- tween applied N fertilizer and (3-amylase activi- ties in on-farm trials when the 24 fields were grouped into zones of N deficiency, optimum N status and excess N in plants using SPAD values (Table 3). Surprisingly, the better N fertilized fields suffered N deficiency for (3-amylase pro- duction. Data indicated that 11 of the 24 grow- ers succeeded in optimizing the N management 498 AGRICULTURAL SCIENCE IN FINLAND Peltonen, J. et ai: Determination of barley nitrogen status with chlorophyll meter for high (famylase in grains Vol. 4: 495-501. without information on plant N status. Plant N status was supraoptimal only in two of 24 spring barley fields surveyed, although the lower dose of N fertilizer was fell within the optimum SPAD zone. The plant N deficiency among 11 fields examined resulted in 19 W.K. lower (3-amylase activity thanfor optimal N status. Based on these data, growers are advised to pursue more vig- orously the N fertilization program in their bar- ley fields to promote (3-amylase production. The data above indicate that the capacity of SPAD values for predicting crop N status and 13- amylase activity at given developmental stages is especially promising. Calculations using Series 1 for the category of adequacy, where the SPAD readings increased from 41 to 45, in- creased P-amylase activity by 143 W.K. and re- quired about 130kg more N ha 1, when compared with the category where the grain yield reached a plateau (data not shown). Thus, to increase P-amylase activity by 10W.K. with N fertilization, 11 kg more N ha 1 was required, which corres- ponded to 40 FIM ha 1 input [calculated using a N (foliar urea) cost of 1.68 FIM kg" I ]. The model developed provides useful information for calculating the price per unit P-amylase activity as a routine part of marketing cv. Kymppi. In Table 3. Comparison of the effect of the intensiveness of N fertilization, based on chlorophyll meter readings, on (1-amylase production in spring barley cv. Kymppi in 24 on-farm trials. Intensiveness of N Mean applied (3-amylase fertilization according N fertilizer activity to SPAD readings rate*rate3 kg ha ' W.K. < Critical SPAD 104(11) 128 Adequate SPAD 94(11) 147 > Optimum SPAD 70 (2) 160 “Number of growers in parentheses addition, based on the knowledge of this physi- cal input-output ratio, growers could express the calculations in economical terms, knowing the current fertilizer cost and price of barley grain and the unit increase in (3-amylase activity. Post-harvest evaluation of P-amylase activity with grain protein concentration Grain protein concentration ofbarley cv. Kymp- pi correlated strongly with (3-amylase activity (Fig. 1), thus confirming findings of Harris and Figure 1. Relationship between grain protein concentration (%) and (3-amylase activity (W.K.) in 1993 and 1994. 499 AGRICULTURAL SCIENCE IN FINLAND Table 4. Suggested critical tissue chlorophyll (SPAD) levels forspring barley cv. Kymppi at PO a in Finland. Traits < Critical’ Adequate' > Optimum 11 Grain yield kg ha 1 37 37-41 46 Grain protein concentration % 41 41-45 48 P-amylase activity W.K. 41 41-45 48 a PO = stage of pollen grains on well-developed stigmatic hairs, coinciding with Zadoks’ GS 52-58. b Critical levels were calculated from Cate-Nelson II models. c Lower and upper limits foradequate SPAD values were developed from Cate-Nelson 111 models. Adequa- cy level can be given if Cate-Nelson 111 models are an improvement in terms of R 2 in respect to Cate- Nelson II models. d Mean values for class of > optimum SPAD, where there were no more increases in grain yield, grain protein concentration and (3-amylase activity. Banasik (1952). There were no significant dif- ferences for levels and slopes of linear regres- sion functions between years (data not shown), indicating that grain protein concentration can be used for post-harvest evaluation of (3-amyl- ase activity in Kymppi within and between sea- sons. A grain protein concentration of 11% indi- cated (3-amylase activity of 152-157 W.K. and at least 14% of grain protein should be reached for (3-amylase activity to exceed 250 W.K. Conclusions In conclusion, the results from this study indic- ated that making accurate N fertilizer recom- mendations based on chlorophyll meter (Minol- ta SPAD-502) readings was economically advant- ageous. Determination of leaf chlorophyll con- tent (SPAD values) using the chlorophyll meter allowed more appropriate fertilizer application recommendations to be made and promote a sub- sequent increase in (3-amylase activity ofa crop. Applying N fertilizer “as needed” resulted in a better physical input-outputratio than with low- er or higher N inputs. Based on the knowledge of the physical input-output ratio, one could ex- press the calculations in economical terms, knowing the current fertilizer cost and price of barley grain and unit increase in (3-amylase act- ivity. Chlorophyll content can be assessed rapidly. The meter’s cost may be prohibitive for many farmers, but not for large growers, consultants and groups including crop management associ- ations. SPAD values for delimiting responsive and non-responsive ranges at pollination (i.e. pollen grains on well-developed stigmatic hairs, GS 52-58) are summarized in Table 4. The ca- pacity of SPAD values to predict barley N status at this stage is especially promising, because supplemental N, applied as foliar sprayed with urea-ammonium-nitrate(Turley and Ching 1986) could easily be applied at this time of develop- ment to increase grain protein concentration and hordein synthesis. The suggested critical SPAD values are suitable under growing conditions where good response of grain yield to N fertil- ization occurs. The SPAD values for optimum grain yield are 4 SPAD units lower than the SPAD values for optimum (3-amylase activity for Kymppi. Grain protein concentrationcould be an effective diagnostic tool for post-harvest evalu- ation of grain (3-amylase activity in Kymppi. 500 AGRICULTURAL SCIIE N C E IN FINLAND Peltonen, J. et cd.: Determination ofbarley nitrogen status with chlorophyll meter for high fi-amylase in grains Vol. 4: 495-501. References AACC 1983. Approved methods of the AACC, Vol. 1,11. Bth Ed. Method 46-11. American Association of Cereal Chemists, Inc., St. Paul, Minnesota, U.S.A. Cate, R.B. & Nelson, L.A. 1971. A simple statistical pro- cedure for partitioning soil test correlation data into two classes. Soil Science Society of America Proceedings 35: 658-660. Pollet, R.H., Pollet, R.F. & Halverson, A.D. 1992. Use of a chlorophyll meter to evaluate the nitrogen status of dryland winter wheat. Communication of Soil Science and Plant Analyse 23: 687-697. Harris, G. & Banasik, O.J. 1952. Effects of environment, variety and season on barley quality. Cereal Chemistry 29: 148-155. Hayter, A.M. & Riggs, T.J. 1973. Environmental and varietal differences in diastatic power for associated char- acteristics of spring barley. Journal of Agricultural Sci- ence, Cambridge 80: 297-230. Helenius, J. 1992. Ohran p-amylaasiaktiivisuuteen tilata- solla vaikuttavat tekijät. University of Helsinki, Depart- ment of Plant Production (M.S. thesis). 45 p. Joy, P. & Peltonen, J. 1993. Breeding implications of the negative correlation between yield and protein con- centration in a winter wheat population. Acta Agriculturae Scandinavica 43: 82-88. Lawlor, D.W., Boyle, F.A., Kendall, A.C. & Keys, A.J. 1987. Nitrate nutrition and temperature effects on wheat: Enzyme composition, nitrate and total amino acid con- tent of leaves. Journal of Experimental Botany 38: 378- 392. Mukula, J. & Rantanen, O. 1989. Climatic risks to the yield and quality of field crops in Finland. VI. Barley 1969- 1986. Annales Agriculturae Fenniae 28: 29-36. Nelson, L.A. & Andersen, R.L. 1977. Partitioning of soil test crop response probability. In: T.R. Peck et al. (eds.) Soil Testing: Correlation and Interpreting the Analytical Results. ASA Special Publication 29. ASA, Madison. Wl. p. 19-38. Peltonen, J. 1992. Ear developmental stage used for tim- ing supplemental nitrogen application to spring wheat. Crop Science 32: 1029-1033, Virtanen, A. & Haggrén, E. 1995. Using a chlorophyll meter to optimize nitrogen fertilizer application for intens- ively-managed small-grain cereals. Journal of Agrono- my and Crop Science 174: 309-318. Turley, R.H. & Ching,T.M. 1986. Storage protein accu- mulation in “Scio” barley seed as affected by late applica- tion of nitrogen. Crop Science 26: 778-782. Waddington, S.R., Cartwright, P.M. & Wall, P.C. 1983. A quantitative scale of spike initial and pistil development in barley and wheat. Annales of Botany 51:119-130. Zadoks, J.C., Chang, T.T. & Konzak, C.F. 1974. A decimal code for the growth stages of cereals. Weed Research 14: 415-421. SELOSTUS Kymppi-ohran (3-amylaasin tuotannon optimointi lehtivihreämittarin avulla Jari Peltonen, Ari Virtanen, Jaakko Helenius, Juha Salopelto, Kari Kiltilä ja Esko Eloranta Kasvintuotannon tarkastuskeskus, Cultor Oy ja Neson Oy Tutkimuksessa määritettiin Kymppi-ohran lehtivihreä- pitoisuuksien raja-arvoja parhaan mahdollisen sadon- muodostuksen ja (3-amylaasi-aktiivisuuden kannalta. Lehtivihreämittaukset tehtiin kasvukaudella 1993 ja 1994 pelto-oloihin tarkoitetulla Minoltan SPAD-502 lehtivihreämittarilla. Mittauksissa käytettiin ylimpiä, täysin kehittyneitä lehtiä. Kenttäkokeina olivat typpi- tasokokeet (0 - 200 kg N ha ')• Lehtivihreämittaukset selittivät (1-amylaasi-aktiivisuutta parhaiten tähkälletu- lovaiheessa (Zadoksin kasvuasteessa 52-58). Taulukos- sa 4 on yhteenveto Kymppi-ohran lehtivihreän raja-ar- voista tässä kasvuasteessa. Lehtivihreän raja-arvo 37 SPAD rajoitti selvästi Kympin sadonmuodostusta. Sa- don menetys oli tällöin 1868 kg ha 1. Jos lehtivihreä- mittaus oli alempi kuin 41 SPAD, laski sadon val- kuaispitoisuus 4,7 prosenttiyksikköä ja (3-amylaasi-aktii- visuus 143 W.K. Lehtivihreän raja-arvot optimisadolle olivat 37-41 SPAD-yksikköä, ja parhaalle mahdolliselle valkuaispitoisuudelle jaP-amylaasi-aktiivisuudelle 41- 45 SPAD-yksikköä. Tutkimus osoitti ohran jyvien fi- amylaasin määrän nousevan voimakkaasti vasta sato-op- timin jälkeen. Sato-optimin jälkeen jokaista kymmentä p-amylaasi-aktiivisuuden yksikköä kohden tarvitaan lisä-N noin 11 kg ha 1. Paras mahdollinen P-amylaasi- aktiivisuus oli keskimäärin 252 W.K. ja vastaava val- kuaispitoisuus 14,2 %. Sadonkorjuun jälkeen voidaan ohran jyvien valkuaispitoisuuden avulla arvioida epä- suorasti jyvien P-amylaasi-aktiivisuutta. 501 AGRICULTURAL SCIENCE IN FINLAND