. - " ‘F ' 3‘7“ sz“*‘ , . J ”43:51, 3.»??? . .Jr _ 51."? . .. ‘é-X- ~3<== _ L511. .4‘ 'raz‘ -‘.‘:.':‘..~._.. "3: .9. _,. ' _ 31—- _ .45": A ‘ fi-h‘. ,4... .. ~. u .'l A " _;-':...L._'.': - ’Jat‘l ‘ -". ::.‘."_.‘-: -.x Jinn." 29 353% :— z‘" :1 - .3; _ ”'2- . ‘ —. 1 A ‘4‘: :J" m. .. loam-".1 .__ -,__‘=.-“—»_.7-:~=;r A. .mv- - :- “ 34‘ ~._;:. weak, , hm, .. up?“ .5...— i 1.7 .1“ “1 M4. ‘ ~ _ "21‘ ' _ ‘c-rkmzc‘fl. “are; ‘ ‘2’ irf " M 'W—T-._... .‘ “a"? .— - .1— .9: .41 :1: . .m?“ 25 “541' var"..- icy: 12:53:33.: 55" w; - J5; 1:— a‘fic‘r-r‘rt :21 gm . y ..1 1 A w. . “" .‘ . ~33: ~ - J. .u—fl 1‘ a I . 3 it”. - . ‘ 11““? 53.2.», . ~< / .17! . 5: 311215 1211123111 1111: M; 2 '11" {1:111 "11111111;( 13:1:5121 11 114fi‘117uhl‘g1 ‘fb .. 1L; ”11.131L11A'fi'i1 111111 1 1" m 1i :1 11:11:11.1 1111 11111;“ 13:111.: 31:: 11111715111117 1 11:1 1;:8 33:31.: -r: ’ 4:9 11 4 ‘1 ‘~ ' . :t‘)'.‘.' I'. {1 ‘. .".~ ‘ .gz'. vV-L. #1:? V A ~ A .. Z ”.5“ h . "jur'fig‘i—g. a wgwfir ...- -: av ‘ 1f: ' #2:... . .fi‘fifl' 5‘ «mi; . film “If; .‘jlféxrfi \ 'I'111‘“r.}(:1“ 11111131111 ' 1 1 ‘1111 1L1:i {11:11:14 7111111111111. 4 11111112: : ~11 :1111I‘1L11L 1111111111111- ":1111: ‘1 «11‘1“ 5:11.111? 1111111113: m 1111111121111: 1113:1315 11,: W11" ' 1 f ' *“T‘I‘Jagg 5* “a: ‘- ‘t _ ‘ . _.d§z£;'é Aug: . _, T ‘1’“ :fl ‘4’”, 332:; a“ “min J l '13 : \2‘ «it my} #3115 @311 ~ :f 3135.. ‘ fig? fir»!- . a. 3 . S .— $.33 3:3- I“? . - ”5;?“ - _‘ . .xi'E-EEv-fl 1, ’ , it: '3? . £35.;— :4"? ,- .,_. ‘1? "$45 wfi: ~ an .40...” W -0... a w .4... EAL ”Wu-I Ifk.‘ .L V- :1 .2: 1.1““? 111% :5?- 1M. 431: t : ”2‘31 . “111‘ :11 1&1“! 11:11!“ ~ - . 5.“: ::‘§‘\$: 43}:- 91.11211 :5 ’1 “3.14;: :3: '1. E11131 1%; 11%;.‘36 “7' {a . :11‘3‘ifitk'3': “:11 " LR. mu :‘t‘x - 1‘ 114;? 151151 119%.??? ,, -u ‘. >9 m,- 1: ‘73:! ‘5‘. . v “,1: flég'fig .. ‘1: 1.x v ‘. M‘- “:51?“ g "3-1: 111%; 1:3: ‘1‘1'I‘é'fi'1tedil u. "1: ”1:21:13 :33 13“,: ‘11 ‘ “1:: y 2:» 5: ,L‘s’ fig? “. 1%“ 13-A- n 1&fi‘1‘t11fi“; a $1?” 11%” 1 1%‘F%:§ifl 5E“. Elkvgifigfisg “w ‘2 ‘ka- 31%;- :5,“ R132". £“mefi u g: 1:: ,, 'u' $1“ i‘r‘ :h Jc 13%. g. y. u {31$ V» 111M 1 1 :ggd-t I : I THESAS £W‘ ‘ Labs—m1! I: --—II _. ‘oo- .v-u‘.‘ - .1 . I V Ema-42 €53? . i High 32‘.” 959.1 5%th i University ) This is to certify that the thesis entitled INFLUENCE OF TEMPERATURE AND IRRADIANCE ON GROWTH AND DEVELOPMENT OF CHRYSANTHEMUM MORIFOLIUM 'BRIGHT GOLDEN ANNE' presented by Meriam G. Karlsson has been accepted towards fulfillment of the requirements for M.S. Horticulture degree in / (6731/49. 5/2124) Major professor Date April 19, 1984 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES .—-:—. RETURNING MATERIALS: Place in book drop to remove this checkout from your record. ElN§§ will be charged if book is returned after the date stamped below. INFLUENCE OF TEMPERATURE.AND IRRADIANCE ON GROWTH AND DEVELOPMENT OF CHRYSANTHEMUM MORIFOLIUM. 'BRIGHT GOLDEN ANNE'. By Mer iam G. Karlsson A.THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER.OF SCIENCE Department of Horticulture 1984 ABSTNMH? INFLWBRJECE‘TEMPEHXHEEIANDIUflEEEANCE<1GCfiKMHH AND DEWEIEWENE’OF(EHQEEWHHEMUMBKXUITKJUM 'BRIGHT(IXIEEIANNE'. By Mer iam G. Kar lsson Manthemum morifolium 'Br ight Golden Anne' plants were grown under 15 combinations of Quantum Flux Density (QFD) , day temperature, and night temperature in a central composite statistical design. Functional relationships between these three environmental factors and subsequent growth were developed. This type of knowledge is necessary for development of growth optimization models. At 20° C temperature, time to flower decreased 30 days when QFD was increased from 50 to 600 umol s'lm’z. Increasing day or night temperature from 14° to 26° delayed flowering. Shoot length increased linearly with day temperature. Total flower area increased as QFD increased or night temperature decreased. Final dry weight at flowering ranged from 4.1 g to 18 9. As QFD increased, partitioning to the roots and leaves decreased while partitioning to the stems and flowers increased. High day temperature increased partitioning to the stems but decreased partitioning to the roots. I would like to thank all those who in different ways have helped me throughout this study. Gratitude is extended to my major professor Dr. Royal D. Heins for help, encouragement and support. Appreciation also goes to the members of my guidance committee: Drs. W. Carlson, C. Cress and J; Flore for their direction and assistance. A special thank you to Mrs. S. Kilyanek for help with data analysis and to Mrs. L. Kent for assistance in the typograhical completion of this thesis. A Financial support for this study was provided in part.by grants from Fred C. Gloeckner Foundation, the American Florists Endowment, and the US Department of Agriculture Small Business Innovation Research program (grant no. 83$BIR-8-0001) in collaboration with Oglevee Computer Systems. GUidance Committee: The paper format was adopted for this thesis in.accordance with departmental and university regulations. Section I is to be submitted to the gggrnal of the American Society for Horticultural Science; and Section II to the Journal of Horticultural Science. iii TABLEWCCNI‘EN'I‘S Page LIST OF TABIE O O O O O O O 0 O O O O O O O 0 v LIST OF. FIGJRE‘S O O O O O O O O O O O O O O O Vii LITERATURE REVIEW Plant Growth Analysis . . . . . . . . . . . . . 1 Statistics of Growth Analysis . . . . . . . . 2 Classical Approach to Growth Analysis . . . . . . 5 Functional Approach to Growth Analysis . . . . . . 7 Available Functions . . . . . . . . . . . . 8 Influence of Irradiance and Temperature on the Development of Chrysanthenum norifolium Ramat. . . . . . . . . 29 Introduction . . . . . . . . . . . . . . . 29 Irradiame O O O O O O O O O O O O O O O 34 Tenperature . . . . . . . . . . . . . . . 41 Partitioning of Dry Matter . . . . . . . . . . 48 Literature Cited . . . . . . . . . . . . . . 50 SECTION I INFLUENCE OF QWI'UM FLUX DENSITY AND TEMPERATURE ON WRING TIME AND PLANT QUALITY OF CI-IRYSANI'I-MM mRIFOLIUM RAMAT. 'BRIGHT GOLDEN ANNE' mstract O O O O O O O O O O O O O O O 56 Material and Methods . . . . . . . . . . . . . 58 Results and Discussion . . . . . . . . . . . . 59 Literature Cited . . . . . . . . . . . . . . 76 SECTION II INFLUENCE OF QUANI‘WI FLUX DENSITY AND TEMPERA‘IURE ON DRY WEIGHT ACCUMUIATION AND PARTITIONING IN CI-IRYSANI'HEMM mRIFOLIUM RAMAT. 'BRIGHT GOLDEN ANNE' Sumnary . . . . . . . . . . . . . . . . 79 Material and Methods . . . . . . . . . . . . . 81 Results and Discussion . . . . . . . . . . . . 83 References . . . . . . . . . . . . . . . . 103 iv Table LIST OF TABLES Page LITERATURE REVIEW Observed values for absolute growth rate, relative growth rate, unit leaf rate, leaf area ratio, shoot-root ratio and leaf area index . . . . . . . . . . . 6 Characteristics of the monomolecular, logistic, Garperz and Richard's functions . . . . . . . . . . . . 26 Characteristics of the mommolecular, logistic, Gouperz and Richard's functions . . . . . . . . . . . . 27 Critical photoperiod for flower bud initiation and flower bud development of 5 varieties of chrysanthenum grown at al6°Ctenperature . . . . . . . . . . . . . 31 Numbers of leaves and bracts initiated before the flower on five cultivars of chrysanthenum grown in long days . . 38 SECTICN I Actual and coded values for treatment combinations used in the central conposite design . . . . . . . . . 65 Influence of QFD, day temperature and night temperature on growth and development of Oirysanthemum morifolium 'Bright when Nine. 0 O O O O O O O O O O O O 66 Regression coefficients for time to flower, shoot length and flower area per plant in Chrysanthermm morifolium 'Bright Golden Anne' . . . . . . . . . . . . . 67 Sierle r values for time to flower, shoot length and flower area per plant in Chrysanthemm morifolium 'Bright Golden Anne' . . . . . . . . . . . . . 68 Time to flower, final shoot length and flower area predicted for ggysanthenum morifolium ' Bright Golden Anne' grown under four different environments . . . . 69 SECTION II Actual and coded values for treatment carbinations used in the central carposite design . . . . . . . . . 89 Table 2. Page Influence of QFD, day and night terrperature on final dry weight accmulation and partitioning in Chrysanthemum mrifolium 'Bright Golden Anne' . . . . . . . . . 90 Regression coefficients for percent root, stem, leaf, and flower dry weights and final total dry weight in Chrysanthemum morifolium 'Bright Golden Anne' . . . . 91 Predicted final total dry weight and proportions parti- tioned to roots, stems, leaves and flowers for Chgsanthermm morifolium 'Bright Golden Anne' . . . . 92 vi LIST OF FIGURES Figure Page LITERATURE REVIEW 1. Exarrples of polynomial curves showing the progression of dry weight and relative growth rate . . . . . . . . 12 2. Exatrples of nonlinear functions showing the progression of dry weight and relative growth rate . . . . . . . 18 3. Two exarrples of Richards curve showing the progression of dry weight and relative growth rate . . . . . . . 21 4. Shape of the Bertalanffy function when m = 2 (the logistic function); 111 = l (the Gonperz function) and m = 0 (the monomolecular function) . . . . . . . . 24 5. Nunber of days from start of short days to flower for a thermopositive, thermonegative and thermozero variety . 32 SECI‘ION I 1. Predicted time to flower as effected by day terrperature, night tenperature, and QFD for Chrysanthenum morifolium 'Bright when me. O 0 O O O O O O O O O O O 70 2. Predicted final shoot length as effected by day tenperature and night tenperature for Chrysanthemm morifolium 'Bright Golden Anne' at 325 1111151 s‘imT. . . 72 3. Predicted flower area per plant as effected by day tenperature, night temperature, and QFD for Chrysanthenum morifolium 'Bright Golden Anne' . . . . . . . . . 74 SECTION II 1. Predicted final plant dry weight as effected by day temperature and QFD for Chrysanthemim morifolium 'Bright Golden Anne' at 200 night tenperature . . . . . . . 93 2. Predicted proportion of dry weight partitioned to the roots as effected by day tenperature and QFD for Qirysanthenum rrorifolium 'Bright Golden Anne' at 20° nighttenperature.............. 95 vii Figure Page 3. Predicted proportion of dry weight partitioned to the stems as effected by day temperature and QFD for Cniysanthenum mrifolium 'Bright Golden Anne' at 20° nighttemperature.............. 97 Predicted proportion of dry weight partitioned to the leaves as effected by day tenperature and QFD for Chrysanthenum morifolium 'Bright Golden Anne' at 20° night temperature. . . . . . . . . . . . . . 99 Predicted proportion of dry weight partitioned to the flowers as effected by day terrperature and QFD for Chrysanthemm morifolium 'Br ight Golden Anne' at 20° nighttenperature. . . . . . . . . . . . . . 101 viii LITERATUREREVIEW LITERATURE REVIEW Literature from two different areas has been reviewed. Plant growth analysis is discussed in the first part and the influence of the environment, primarily irradiance and temperature on growth and development of Chrysanthemum morifolium Ramat. is the tcpic for the last part of this literature review. Plant Growth Analysis Various procedures are used to compare plant growth and development. Many of the procedures used in growth analysis were first studied and defined at the beginning of this century as rates and ratios (30,31,70) ; these calculated estimators of population parameters will be called statistics in this review (63). When the statistics are estimated as a mean value over the time period between data collection, the calculation method is referred to as the classical approach to growth analysis. When the statistics are derived from fitting mathematical functions to the raw data (40,42), the calculation method is referred to as the functional approach. Several statistics are described below and some typical values are presented in Table l. The classical and the functional approaches are then discussed followed by mathematical functions typically used in the functional approach. Statistics of Growth Analysis Growth can be described as a function of time: W = f (t) (1) where W is total plant dry weight at time t (l6,30,40,42,59). The absolute growth rate (G) is given by the derivative of this function: G = dW/dt (2) Absolute growth rate has often been observed to be approximately proportional to the size of the plant (15,16,59). Therefore absolute growth rate isn't necessarily the best way to describe a plant's physiological performance. Dry matter gain per unit plant weight is another way to express the production efficiency. This statistic is called the relative growth rate (RGR) and is the absolute growth rate divided by the existing weight (6,9,15,16,30,31,40,42,59): RGR = (dW/dt) x (l/W) (3) Also since, by definition, d(ln W)/dt = (dW/dt) X (l/W) (4) the first derivative of any total dry weight function expressed as the natural logarithm of total dry weight automatically gives RGR. The mean relative growth rate (RGR) between two times (T1 and T2) can be expressed: fi=<1nw2-1nwl)/(T2-Tl) (5) Equation 3 gives instantaneous values of RGR. Hunt (42) has shown that RGR often changes smoothly over time and this drift can often be followed by deriving mean relative growth rates between harvest intervals. As the harvest intervals become shorter the mean relative growth rate gives better and better estimates of instantaneous RGR. The RGR is useful for growth rate comparisons between experiments and species. But this method implies that all parts of the plant are equally efficient in producing new dry matter. In most plants the leaves are the main site for photosynthesis and Briggs et al. (10) found that the Weekly increase in total plant dry weight per unit leaf area for a particular species and set of environmental conditions is rather constant throughout plant development. The net weight gain per unit leaf area seems to be an appropiate index for plant assimilation efficiency. This weight gain has been called Unit Leaf Rate (ULR) (10) and the instantaneous value can be expressed: ULR = (l/LA) x (dW/dt) (6) where LA is the plant total leaf area (9,16,30,40,42). Sometimes the ULR is called Net Assimilation Rate (NAR) (30,40,42,67,70). Before the term ULR was introduced by Briggs et al. (9) the only existing name for this statistic was the German word 'Assimilationenergie' and since NAR can be confused with the term apparent assimilation, which relates to the photoreduction of carbon dioxide, the term ULR is preferred (30). The Leaf Area Ratio (LAR) is the ratio between leaf area and total dry weight : LAR = LA/W (7) LAR can be broken into two parts, specific leaf area (SLA) and leaf weight ratio (LWR). SLA is the leaf area divided by leaf weight (LA/Lw) and is a measurement of leaf density or relative leaf thickness (42). Plant 'leafiness' can either be expressed on an area/weight basis (SLA) or on a weight/weight basis (IW/W) as in LWR (42)- The RGR can be expressed with the help of ULR and LAR (9,16,30,40,42) : (l/W) x (dW/dT) ((l/LA) x (dW/dT)) x(LA/W) (8) RGR ULR x LAR In some experimental analysis the relationship between shoot dry weight and root dry weight is of interest. The statistics are simple ratios (16,42) : Rw/Sw 0r Sw/Rw (9) ULR is not appropiate when a population of plants is studied. This is because spacing between plants must be taken into account and measurements of 'leafiness' in relation to land area gives more information about a whole crops potential productivity. This ratio between total leaf area and the occupied land area (P) is called leaf area index (LAI) (30,42) Only the most common ratios and rates in plant growth analysis have been discussed here but many others have been defined (16,30,42). Some statistic values observed in plant growth analysis are presented in Table l. gassical Approach to Growth Analysis The ratios and rates mentioned above were traditionally calculated from the raw data without further attempts to find underlying mathematical functions (16). This procedure of calculation on raw data is referred to as the classical approach to plant growth analysis. The main advantage of the classical approach is the ease with which rates and ratios can be calculated. However assumptions must often be made. For example, when calculating mean values of quantities like ULR, weight and leaf area are assumed to be linearly related over the time period (30,40,42,67). This isn't necessarily the . case for fast growing plants or long harvest intervals (40). While frequent sampling is necessary for the functional approach, the classical method can be used with a small number of sampling periods (40,42,67). Since plant dry weight measurements are 6 lhhle 1. Observed values for absolute growth rate, relative growth rate, unit leaf rate, leaf area ratio, shoot-root ratio and leaf area index. Calculation Range of Statistic method typical values Unit Species References A:::l::‘ §%' 0.01 g den"1 Hglgzsa Hunt (1978) rate 1.9 Haize Hunt (1978) 0.01 - 10.26 Helianthus annuus Evans (1972) Relative . -l g; l 0.06 - 0.16 day Phalaris g::::h dt . w tuberosa Hillians (1946) parviflora Evans (.1972) 0.088 - 0.20 Helianthus Evans (1972) annuus 0.262 - 0.482 Pigweed Potter. Jones (1977) 0.39 Poa annuus Hunt (1978) Unit 1 du -2 -1 ~-.— 5.6 - 10.2 g m day Chrysanthemum 1::: LA dt morifolium Hughes (1973b) 2.07 - 4.72' Impatiens perviflora Evans (1972) 8.47 Helianthus annuus Hunt (1978) 9.77 Apple Haggs (1960) -21.4 - 17.9 Maize Briggs et al. (19206) L." LA 0 o 004 2 " cm: 1: 1: area .__ - . m g s ep us r.“ U chimnsis Evans (1972) 0.0044 Pinus syl- Hunt (1978) vestris 0.0006 - 0.022 Maize Briggs et al (1920a) 0.0177 Helianthus annuus Hunt (1978) 0.01 - 0.02 Chrysanthemum Inrifolium Hughes (1973b) swim: 5" 2 03 2 36 u 1i th rat 0 ‘- . - . -- e an us Ru annuus Evans (1972) 3 - 5 Inpatiens parviflora Evans (1972) 4.17 - 6.17 Helianthus Evans (1972) debilis 0.48 Sugarbeet Milthorpe.Moorby (1979) Leaf area LA 0 - 3 -- Sugarbeet Hunt (1982) "'4‘" 15‘ o . 3 Wheat Hunt (1982) -0.2 - 8.84 wheat Austin et aL (1980) 2.2 . 12.6 Chrysanthemum Acock et a1.(l978) morifoliue ————————~'~. ______._...._.— destructive, a plant can only be sampled once. This problem has been handled for years in the classical approach by pairing plants. The largest plant in harvest one is paired with the largest plant in harvest two etc. (16,30,40,42). Differences between plants are reduced with this method and the experimental error is primarily random. Rates estimated using the classical approach are sensitive to sampling errors and environmental variations. Therefore the overall trend might be hard to interpret (16). Curve fitting as described below in the functional approach often makes it easier to follow both the development of the plant and the statistics of interest (15,16,40,42,59) . Functional Approach to Growth Analysis Fitting functions to experimental data using regression analysis is referred to as the functional approach to plant growth analysis (16,28). Three statistical requirements must be fulfilled for regression analysis to be valid when fitting functions to growth data. The independent variable (X) should be measured without errors, the distribution of measured Y values at each X should be normal, and the variance of Y at each X should be uniform and not change throughout the analysis (28,42). Time is usually the independent variable and can be virtually measured without errors. But the second and third requirements for regression analysis sometimes cause problems. The conventional method to satisfy the last two requirements is to transform the data (28) by taking the natural logarithm (base e) of each datum point. Transformation using any other base would be equally efficient to fulfill the statistical requirements (42). The functional approach has many advantages and computers have made the method possible to use. Complicated mathematical equations once avoided can now be quickly and accurately calculated (40,41,42). Experimental data contain random errors and a fitted function generally smooths these variations to give a growth curve free from large fluctuations (16,42). Each point on the curve contains information from all sampling occasions (40,41,42) and the model with the information condensed into a few parameters often become more important to the experimenter than the data from which it was derived (42). Available Emotions The two types of functions mainly used in the functional approach to plant growth analysis are polynomial functions and asymptotic functions. Polgaomial functions have been extensively used in plant growth analysis. This is not due to any biological significance, but rather that they are a simple kind of mathematical function (15). Polynomial functions which have linear parameters or parameters which can be transformed to a linear form can be fitted to data by exact and well defined multiple regression techniques (28,63). A polynomial has the form: y=a+b1x+b2x2+... +ann (11) The coefficients 'a, bl . . . bn' are estimated in the regression analysis, and the highest power of the independent variable determine the name of the polynomial (15,16,42,59). The first order polynomial or 'linear regression' have the following form when applied to total plant dry weight (15,16,42). W=a+bT (12) To fulfill the statistical requirements mentioned earlier concerning regression, transformation before curve fitting to natural logarithms is often done. The first order polynomial in exponential form will be: an=a+bT (13) The absolute growth rate (dW/dT) is given by the derivative of equation 12 (42,59) : G = dW/dT = b (14) If the natural logarithm is used as in equation 13 the derivative calculates RGR: b = RGR = (d(ln W))/dT = (l/W) x (dW/dT) (15) Coefficient 'a' implies the size of the growing system at the time chosen to be zero, and 'b' is the rate of increase in W (absolute 10 growth) or 1n W (RGR). A constantly increasing W will be the result of a positive 'b' value and decreasing W with a negative 'b' value. When 'b' is zero, W will be equal to 'a', see Figure 1a. The first order polynomials are appropiate functions when growth occurs by equal cell division at regular intervals. But meristematic tissues cannot keep on dividing for long time periods without cell differentiation. The use of first order polynomials is therefore limited to short periods of growth in young plants or parts of plants (42). The second order polynomial has the form: W(or1nW) =a+b1T+b2T2 (16) As in the first order polynomial the derivative of equation 16 will give the absolute growth rate when applied to untransformed data and KR for transformed data. dW/dT (or (l/W) x (dW/dT)) = bl + 2b2T (17) Coefficient 'a' is the size when T equals zero, 'bl' represent SIOWth rate at time zero and 'b2' the amount of curvature or rate of change of the growth rate (42). The second derivative of equation 16 is: d2w/d'r2 = 2 b2 (18) and this stands for acceleration or the rate of change of the rate of 11 change of W. A sample of second order curves is shown in Figure lb. The second order polynomial is a growth curve where the growth rate always will be a first order function (Figure lb). This might be a limitation, since no inflections in the growth data can be illustrated. But it is a simple growth curve and good fits are often obtained for at least parts of a growing process (42). An increase from second to third order polynomial will give the following equation: W(oran) =a+blT+b2'I'2+b3'I3 (19) The growth rates of this function are given by dW/dT (or (1m) x (dW/dT)) = bl + 2sz + 3b3T2 (20) The cofficient 'a' is as in all polynomials the starting size of the system (42). Growth rate at time zero is given by the coefficient 'bl'. A third order polynomial can take many different shapes and a few examples are shown in Figure 1c. This polynomial can be considered as a function for relationships which curve in one direction or change curvilinearity over time (42). Polynomials with higher order than three have great flexibility and can describe many biological processes; however the coefficients don't have any biological significance and the functions are just empirical equations. This is one limitation for use of higher order polynomials. Another possible limitation is the size of the computer Figure 1. 12 Exanples of polynomial curves showing the progression of dry weight (——) and relative growth rate (-——‘); a) first order polynomials; b) second order polynomials; and c) third order polynomials. 8a.. 5395 2580: O V _ It I —_______- A _ _ _ _ u _ . A 29.3 to 32 5395 2550: Time A Tlmo 32 5:65 2530: /\ Time l4 facility. As the number of coefficients increases, the coefficient's numerical value usually decrease and more memory space is required for precision. There also is a risk for overfitting with higher order polynomials, since a function exactly fitting every point can be developed (16,42). From a growth analysis stand point this is not desirable. No 'smoothing' of the data has been done and the overall trend cannot readily be seen (40,42). Asynptotic functions are nonlinear in the parameters by means of multiplication, division or exponentiation with each other (16,28,42). Because of the nonlinear nature there is no direct method ; for parameter estimations. Arbitrary starting values are usually assigned to all or some of the parameters and with this starting equation the best possible statistics are calculated through several iterations. Good calculating facilities are necessary for fitting of nonlinear functions and for many years this has been a limiting factor. Only during recent years with the development of high capacity computers have the asymptotic functions become reasonable to use in growth analysis (l6,28,40,41,42). When equation 15 is integrated the result is the so called mntial equation (42,59) : w = a ebT (21) where coefficient 'a' is the initial system size at the beg inning of the study and 'b' is the rate of increase in growth (42,59). The monomolecular function was developed to illustrate the progression of a first order chemical reaction (29,42,48,59). With the 15 notations used here for growth analysis the monomolecular function has the form (42): w (or 1n W) = a(1 - be'CT) (22) This function is constantly increasing from the point 'a(1-b)' at time zero (28) and has no point of inflection (28,42) as shown in Figure 2a. Coefficient 'a' is the asymptotic value which determines the range of the dependent axis, 'b' is a measure of where the intercept will occur and coefficient 'c' is a rate constant controlling the spread along the independent axis (42,59). From equation 22 the rate of growth is given by the derivative (42) : dW/dT = abc e‘CT (23) 1/w x dW/dT = (bc e'CT) / (1 - b e'CT) (24) The growth rate is proportional to the amount of growth yet to occur (28,42,48,59) and is continuously decreasing (59) see Figure 2a. The monomolecular growth function has primarily been used for fitting data from later parts of plant growth (28,59). A growth function where the rate of growth is proportional to the present size and to some assumed final size is called the 1gistic l6 equation (28,29,59), since the original use of this function was for an autocatalytic monomolecular reaction the name autocatalytic is sometimes used (28,29,42,58,59). The form of the logistic function is: w (or 1n W) = a/(l + b e-C'I) (25) The growth curve is S-shaped with an inflection at the point W = a/2 (16,29,59). This inflection point divides the curve into two parts which have different directions but otherwise are identical (59). At time zero W is 'a/(l+b)' and the function is asymptotic to W = 0 and W = a (29,42,48,59). The constants 'a', 'b' and 'c' have the same biological significance as in the monomolecular function (42,48,59). Growth rate or the. slope can be calculated from the derivative of equation 25: dW/dT = (abc e-C’I‘) / (1 + b e-CT12 (26) 1/w x dW/dT = (bc e-CT) / (1 + b e-CT) (27) The logistic function is a relatively simple asymptotic function and it often gives a good fit to growth data. Because of this the logistic function has been popular in plant growth analysis (42,59). Figure 2b illustrates the logistic function and its slope. 17 A third growth function with three coefficients often used is the €0er function. The three coefficients are arranged in a double exponent (16,28,29,42,58,59): -CT W(oran) =ae"'be (28) The final size 'a' is approached asymptotically and W equals zero when T = -oo (59). At the size 'a/e' (0.3679 a) the point of inflection occurs (28,29,42,59). Many growth data have their maximal growth rates somewhere between 'a/3' and 'a/2', and the Gomperz function will reproduce these growth processes well (59). As in the monomolecular and the logistic functions coefficient 'b' is a measure of initial system size and 'c' is a rate constant (59). Derivation of equation 28 gives the rate of growth (42): -cT dW/dT = abc e‘cT‘b e (29) l/W x dW/dT = bc e‘CT (30) The Gomperz function was developed for work with animals and population studies (16,28,29,59). In plant growth analysis it has often been adapted to growth of parts of plants, especially to leaf growth data (42,58,59). Figure 2c gives a graphical representation of equation 28, 29 and 30. The Richards function is a four parameter function and was 18 Figure 2. Ekamples of nonlinear functions showing the progression of dry weight (---) and relative growth rate (--w-); a) monomolecular function; b) logistic function; c) Gomperz function. one. 5905 033.3... 0 ¢ _ _ . . C a C \ ~ \ $203 >5 38 526.5 025.2... 0 _ _ e \ Ego; to 39 525.5 025.3. £203 to Time 20 introduced by F.J. Richards in 1959 (28,42,58,59). Its form is shown in equation 31 and the derivatives in equation 32 and 33. W (or In W) = a(1 i e(b'CT))-l/d (31) dW/dT = (ac e 1 the result will be the Gomperz function (28,58,59). The curve shape will continuously change from 21 Figure 3. Two exanples of Richards curve showing the progression of dry weight (—-—) and relative growth rate (.——-) . 32 £326 2596: O Time 23 monomolecular into Gomperz form when the 'm' value goes from 0 to l, and from Gomperz into autocatalytic form when 'm' increases from 1 to 2 (59) , see Figure 4. Where the inflection point is on the growth curve depends on the size of 'm'. Larger values of 'm' will move the inflection point to the later parts of growth development. In Richards function (equation 31) the coefficient 'd' controls where the inflection point will occur on the growth curve (42). The other coefficients have the same biological significance as in the three earlier mentioned growth functions used in nonlinear growth analysis. Richards function has lately become popular in growth analysis. It gives a good fit to many plant growth data, especially when parts of plants are studied. In whole plant studies however the first and the last part of the develcpment sometimes cause problems, since Richards function doesn't seem to reproduce the growth pattern well at these developmental stages (42). Another problem, which might be encountered is the increased difficulty of estimating and finding starting values for four instead of three coefficients. Tables 2 and 3 are a summary of some characteristics for the growth functions discussed here. The term modeling is now frequently used for studies applying the functional approach to data analysis. Thornley (65) described a model as a set of mathematical equations, which quantitatively represent the assumptions made about a studied system. When equations are fitted to experimental data the model is empirical. This type of modeling is most suitable as a first approach to a problem. It might be possible with this model as a basis, to look at the mechanism behind the responses and make a so called mechanistic model (65). 24. Figure 4. Shape of the Bertalanffy function when m = 2 (the logistic function); m = l (the Gomperz function) and when m = 0 (the monomolecular function) . £26; to Time 26 £822, a 21 an: 21:938.: 1 3 must; m .m a nice onion u z Namaeoo has .m. .1 2+5} rowers: ... 3 3333 one: a Ani—Vo Awouoaipva u z co—aumpeeocoz z 5 e52. 3" e o u u coFHom—ucm am 3 mo o=—e> copaezam =o.au==n .mcowuoeam m.ocu;u_¢ one ~goaeo¢ .u_um.mo— .gopauo—oeoeoe we» we mopam_gmuomeo;u .N o—aeh 2'7 0 one.- node a i. i one». .~+e\—v he a + penance eoae huieAIhoi A ee+~v leeae N POI he AhoIAdevv A . a eiuveve. u 9.3. ill 1 . a eu : u A e A (IIIIII. he); vxn h .3 + Aq+axuv bone + huiawee Au+exuvi beta + v aunaeue eeuesuwu one A eu e he: i i A eun- emcee be 99-03 Hoozioev haleAJ—bi 8 3+. . anlumm «ahulea+~uiil «Aficiea+~v equeuuoa t- aviatoizeia tux. sou-mu— hurumm AholakoIOAiuveveuae antenna humane—alone: euee euev euee use eeeuoueeeuai. eve-nomad: genome-.93.! so: u .eeoquuesu e.sueau«u one eyes-ea .uqueuaeu .ueueee—eeeeel one no euuueuueuueueao..n euaeh 28 The main problem with a functional approach is to decide which function is most suitable to use for the growth analysis in question (l6,40,42,67). Classical estimated parameters often give an indication of the overall growth trends and the form of the underlying growth functions can be distinguished easier (30,42). A combination of classical and functional methods is necessary for successful growth analysis. Influence of Irradiance and Tertperature on the Development of Chrysanthemum morifolim Ramat. Msanthemum moiifolium Ramat. is one of the most important crops grown in commercial greenhouses today (2). This review, will emphasize how irradiance and temperature influence the growth and development of Chrysanthemums grown as pot plants. The influence of irradiance and daylength on time to flower and plant appearance (height, number of leaves and flowers, flower diameter etc.) will be described, followed by the influence of different day and night temperatures on time of development and final plant appearance. Partitioning of dry matter will be discussed in the last part of the literature review. Introduction Ci'irysapthemum mogfolium Ramat. has been classified as a short day (SD) plant (14,24,62). The critical photoperiod was reported in 1939 to be 14 1/2 hours (9 1/2 hours darkness) (54). Later Post (55) discovered that 14 1/2 hours was the critical photoperiod for flower bud initiation and that the critical photoperiod for development of the flower buds was 13 1/2 hours (10 1/2 hours darkness). The time necessary for flower development after start of short days varies with cultivar; cultivars are classified into response groups based on the number of weeks from start of 80 to flower (46). Response groups vary from 6 weeks to 15 weeks (3). 29 30 Doorenbos and Kofranek (27) found flower initiation to initially occur at the same rate after the start of SD in early (9 weeks) and late cultivars (14 weeks) but subsequent flower develcpment was slower in the late varieties. Critical daylength was shorter for late cultivars than early cultivars (32). Langhans (46) published the critical daylengths for different response groups after data by Cathey (15) (Table 4). Flower development in Chrysanthemum is affected by both photoperiod and temperature. In 6 to 7 week response group cultivars, temperature seemed to be the dominating factor, while daylength was more important for the development of a longer response group (47). Cathey (12) divided Chrysanthemums into three different groups based on their response to temperature. Cultivars that flowered in a temperature range of 100 to 27° with the fastest development at 16° and only slight delay at 10° and 27° were called thegmozero cultivars. When a minimum temperature of 160 was necessary for initiation of flower buds, the cultivars were called thepmopositive. In this group temperatures below 160 inhibited initiation and development of flower buds. The third group was called thermonegative, since temperatures above 160 inhibited flowering. Flower buds in this group were initiated at higher temperatures but failed to develop. Figure 5 shows the response of temperature on time to flowering for a thermozero, a thermonegative and a thermopositive cultivar. When the cultivar Lilian Doty was grown at 130, 170 and 210, SD only induced flowering under 21°. The plants remained vegetative at the lower temperatures even with SD (60). Post and Lacey (56) showed that high temperatures during SD also can delay flowering. It appears that bud initiation and development under SD is 31 Table 4. Critical photoperiod for flower bud initiation and flower bud dexelopment of 5 varieties of Chrysanthemums grown at a 16 C temperature (from Langhans, 1964 after data from Cathey, 1954). Critical photoperiod (hrs) Response Flower bud Flower bud Variety group initiation development White Wonder 6 16 13 3/4 Pristine 8 15 1/4 12 Encore 10 14 1/2 12 Fortune 12 13 ' 12 Snow 15 11 10 Figure 5. 32 Number of days from start of short days to flower for a thermopositive, thermonegative and thermozero variety planted in early January from stock plants kept at 16°. The plants were grown in a night temperature range from 10 to 27°. (Redrawn from Machin and Scope 1978 after data from Cathey 1954a) . < 223 34 dependent on temperature and optimum temperature varies with cultivar (47). I A partially differentiated shoot apex where complete development is arrested is called a crown bud (14). This kind of bud has strap-shaped leaves beneath it, while a normal terminal bud has lobed leaves below it (14). Flowering is often described with criteria like number of developed leaves, number of days to visible bud or days to anthesis and a measure for vegetative growth often used is internode length (14) . Irradiance Schwabe (62) found that the time required for flower bud initiation and time to flower under short days to be affected by seasonal changes in Quantum Flux Density (QFD). As irradiance increased, the transition to reproductive development as indicated by earlier appearance of flower buds and less number of leaves below the bud, began earlier even though all plants were under short days (19,62). Hughes (34) experimenting with different daylengths and irradiance found vegetative growth to be primarily dependent on total daily irradiance, irrespective of photoperiod (8 or 12 hour). Fastest flower development occured under the conditions of highest irradiance (95 J cm'zd"1) and 8 hours daylength. This irradiance corresponds to 150 umol s"’lm"2 during the 8 hours light span. An almost linear relationship between total dry weight and irradiance at constant daylength was observed (34). The cultivar 'Br ight Golden Anne' flowered after 70 short days 35 when grown under either 125 or 250 J cm‘2 8-hr d‘1 (200 or 400 umol s"]-m"2 for 8 hours) (36). At 31 and 63 J cm'z 8-hr d‘1 (50 and 100 umol s'lm“2 for 8 hours) flowering occured after 94 and 87 short days respectively. Cockshull and Hughes (23) concluded that an irradiance of 125 J cm‘2 for 8 hours per day (200 pmol s‘lm‘z, 8 hr dTI), was adequate for normal flower development. Transferring plants from an irradiance of 63 to 125 J cm'2 d‘1 (from ca. 100 to 200 umol s"lm"2 on an 8 hour basis) during the first two weeks of short days hasten flower initiation and decreased time to flowering compared to plants grown continuously at 63 J cm‘zd‘l (24). The effect on flower development was greatest when the high irradiance was provided at the beginning of short days; two weeks at 125 J cm'zd'l (ca. 200 umol s‘lm‘z, 8 hr d‘l) were more efficient (faster flower initiation and development) than one week. A low irradiance (31 J cm‘zd'l, corresponding to ca. 50 umol s'lm'z for 8 hr d'l) after the two initial weeks at 125 J cm"2 d‘1 for 8 hr d-l did not stop further development of flowers but the final flower quality was poor (retarded floret initiation and a large variability in flower development) due to the low average irradiance of 47 J cm’zd‘l (75 umol s‘lm’2 for 8 hr d'l) during the whole short day period (24). Plants grown continuously at 63 J cm"2d‘l had a more variable development than plants under 125 J cm‘zd‘l (24,36). Cockshull and Hughes (23) showed this increased variability to be due to variable flower initiation under the lower light at the beginning of short days. Chrysanthemums under a constant irrad iance of 125 J cm‘zd‘l developed similar to plants receiving the same total irradiance but given alternately as 31 and 219 J cm’zd‘l (50 and 350 umol s’lm‘z for 8 36 hr d'l) (24). This similarity is not surprising as the reaction of light in photosynthesis is primarily photochemical (50). The amount of photosynthetically active quanta absorbed will determine photosynthesis and the dry matter production would be expected to be similar at the same average QFD (50). Stepped irradiance was studied by Hughes and Cockshull (37) in an effort to resemble diurnal irradiation with higher intensities at noon and lower intensities at the beginning and end of a day. Morphology and growth in the range from 31 to 250 J cm‘zd'l (50 to 400 umol s‘lm‘z, 8 hr d'l) was found to be a function of total daily irradiance rather than to changing irradiance during the day. Schawbe (62) concluded that the seasonal differences in time to flower was correlated with changes in irradiance. However no seasonal changes in leaf number were observed. When Cockshull and Hughes (24) grew plants under 63 and 125 J cm‘z d‘1 (100 and 200 (111101 s'lm‘z, 8 hr d‘l) they found a higher leaf number at the lower irradiance. Similar results have been reported by Hughes and Cockshull (36); 15 leaves were formed at 31 J era-2 d‘1 (50 umol s-lm-Z, 8 hr d'l), 10 at 63 J cm‘2 d‘1 (100 umol s’lm‘z) and 7-8 leaves at) 125 and 250 J cm-2 d"1 (200 and 400 pmol s‘lm‘z, 8 hr d‘l). The shoot height was shorter in the highest and the lowest irradiance (15.8 - 21.7 cm) than in the middle two irradiance levels (16.6 - 27.4 cm)(36). Supplemental lighting of flowering pot plants during low light conditions often result in improved quality (11). Lighting at 5 W ft":2 for 10 hr d-1 (270 pmol s‘lm'z, 10 hr 671) of 6-inch pot Chrysanthemums during dark winter months resulted in plants with increased flower 37 number (up to 5 flowers/plant), dry weight (2-4 grams/plant) and stem diameter..An increase in plant height (13 - 39 % depending on cultivar) also occured under the increased irradiance (ll). Eyen under continuous long days, Chrysanthemums will eventually initiate flower buds. The number of leaves initiated under long day conditions varied both with variety and time of year (17,47). However when the cultivars were ranked by leaf number, their relative positions were always the same as shown in Table 5. Flower initiation in long days was related to an ageing process of the apical meristem (17,18,47). The time necessary for this process was influenced by environmental factors. Cockshull (19) found that under continuous irradiance (24 hours a day) fewer leaves were initiated at 120 W m-2 (550 umol s'lm'z) in the cultivars 'Polaris' and 'Bright Golden Anne' prior to flower bud initiation than on plants grown under 7.5 W m"2 (35 umol s'lm‘z). Above 60 W m‘”2 (280 umol s‘lm’z) the leaf number approached a minimum and the rate of leaf initiation increased with irradiance reaching a maximum above 60 W m‘z. Temperatures in the range 16 to 280 had little effect on time to flower initiation in continuous light (17,18). Cockshull and Hughes (23) found the number of initiated florets per flower to be higher when. plants were grown at 375 J cmfzd'l (600 umol s"lm'2 for 8 hr h‘l), than when grown at 31 J cm‘zd‘l (50 umol s‘lm‘z, 8 hr d'l). The irrad iance level between the 15th to let short days was the most important.in influencing floret number. Total dry weight increase was approximately proportional to increasing irradiance up to 125 J anQd‘l (200 umcl s"’lm"2 for 8 hr d'l), while a linear effect of irradiance in the range 63 to 250 38 Table 5. Numbers of leaves and bracts initiated before the flower on five cultivars of Chrysanthemum grown in long days (Natural daylength plus 5 h night break). (After Cockshull, l974). Date of Planting Cultivar 6.l3.73 l0.lO.73 5.29.74 Average Tuneful 45.3 90.3 56.9 64.2 Gold Crystal 44.0 69.2 49.5 54.2 Polaris 33.5 56.l 40.8 43.5 Bluechip 29.9 48.4 33.8 35.4 Bright Golden Anne 20.3 34.3 18.4 24.3 39 J cm'zd‘l (100 to 400 umol s‘lm‘z, 8 hr d’l) on flower dry weight was observed (24,37,38,39). Carbon dioxide enrichment had a greater effect on flower dry weight than on total dry weight. Hughes and Cockshull (36) explained this to faster flower development and greater partitioning of dry matter to the flowers. lowering irradiance (from 375 to 125 J cm‘zd'”l or from 125 to 31 J cm'zd‘l) during any stage of the short day period generally reduced both total and flower dry weight (23). Higher irradiance (125 or 375 J cmfzd‘l) during the first four weeks of short days didn‘t result in any detectable increased total dry weight at time of flowering if plants were shifted tora Lower irradiance~during the final 6 weeks of development. Transfers after five weeks of short days to higher irradiance from lower irrad iance levels produced a significant increase in total dry weight. After five weeks maximum leaf area had developed and a higher irradiance could be used more efficiently by the plants to produce dry matter (23). Only a small difference in total dry weight production.has been detected when the same total irradiance (in a range up to 250 J cm‘zd‘l) was given during a day, irrespective of daily timing (34). For example, the average daily irradiance could be given in a rising and falling diurnal cycle (37); by alternating days at high and low irradiance (24) or by exposing plants to different irradiance with inversely compensating daylengths (34). There did not appear to be a requirement for a certain leaf number or area before flower initiation could occur (23). However flower initiation was delayed under Low' irradiance (31 and 63 J cm'“2 8-hr d‘l), and the number of leaves formed often was larger 40 compared to plants grown under 250 J cm"2 8-hr d‘l). Total leaf area per plant was similar for all light treatments and maximum leaf area was developed by the end of six to seven weeks of short days (23). Hughes and Cockshull (36) found a smaller total leaf area under low irradiance (31 J cm"2 8-hr d'l) and C02 concentration (325 11:1 1'1), than under 125 J cm""2 8-hr d"'1 and 900 111 1‘1 C02- The higher irradiance and C02 combinations generally had a larger leaf area, but no consistent pattern could be distinguished. Unit Leaf Rate (ULR) increased with increasing irradiance (31 - 250 J cm-Z 8-hr d'l) and C02 levels (325 - 600 -ul 1'1) from 0.08 to 0.5 mg cm"2 d"l when the plants were 20 days old (36). When this experiment was repeated with plants initially smaller, the ULR was higher for corresponding combinations of irradiance and C02. A downward trend for ULR occurs on growing and developing plants since intraplant shading increases as the plant gets larger (36). Leaf Area Ratio (LAR) decreased with increasing light and flower development (36) . The Relative Growth Rate (KER) when the plants were 40 days old decreased from 0.042 d'1 under a 12 hour photoperiod with a high irradiance (33 J m‘zs‘l, corresponding to ca. 150 -1.1mol,s"lm"2 for 12 hr d'l) to 0.035 (3’1 under an 8 hour photoperiod with a low irradiance (22 J m"23'l or 100 umol s‘lm‘z, 8 hr d‘l) (34). Under the same conditions, LAR increased from 90 to 160 cng'“1 (0.0009 - 0.016 ng’l), while ULR decreased from 0.39 to 0.2 mg cm‘zd’l (3.9 - 2 g m‘zd‘l) (34). Plants grown in daylengths of 8 or 12 hours didn't show any difference in specific respiration rates (34). However, there was a 41 decrease over the dark period and the overall period of development. Based on total dry weight, mature flowers were found to have the same respiration rate as the rest of the plant (34). Hughes and Tsjuita (39) found Leaf Weight Ratio (LWR) to be relatively unaffected by irradiance. However Hughes and Cockshull (36) found that LWR was greater on Chrysanthemums grown under low irradiances (31 - 63 J cm‘2 8-hr d’l) than at higher irradiances (125 - 250 J cm"2 8-hr d'l). Tenpegature A night temperature of 27° hasten bud formation and plants had a lower percentage of blind shoots (shoots failing to form flower buds) than when plants were exposed to a night temperature of 10° C (54). A combination of low night and high day temperature (10° and 21°C) produced more flower buds than the reciprocal combination (21° night and 10° day temperature). Under low irradiances and high temperature conditions the initiation of flower buds was poorer (more blind shoots) than under higher irradiances (54). The cultivar 'Sea Gull' formed flower buds under night temperatures from 16° to 32°, but at an average night temperature of 32° C the buds failed to develop into flowers. At 30° C, flowering was delayed 11 days compared to plants grown under cooler night temperatures (33). Cathey (12) studied temperature effects on bud initiation and flower development in Chrysanthemum. His results showed delayed bud initiation at temperatures either above or below 16° C. The longer the 42 high or low temperatures were maintained during bud initiation the greater the delay (12). Samman and Langhans (61) found low night temperature during the initiation phase to give the greatest delay in flower development with a maximum delay at 4.5°. Similar results have been reported by Vince (68). Low night temperature (4.5° — 10°) during initiation until the bud was visible delayed flowering up to 100 days and at the lowest temperature (4.50) most cultivars failed to flower. Low night temperature (4.50 - 10°) after the visible bud stage had little or no effect on flower development. However an interaction between night temperature and light intensity was observed. A reduced QFD (1/3 of average natural daylight during fall and winter in England) during the short day period delayed flowering ”considerably. Several cultivars were investigated in a Dutch study to find optimum night temperature for flower development (43). The best development occured at night temperatures between 17° and 21° C. Cathey (13) reported 16° to be the best temperature for growth of Chrysanthemums. The difference in results may be due to cultivar differences or the difference in latitude between the two places where the experiments were conducted (Holland at 52° north and Ithaca, New York at 43° north) (43). At Ithaca the experiment started in January and was conducted under natural days, in Holland short days (15 hours dark) were provided with black cloth. At a temperature combination of 22° day and 18° night, Chrysanthemums flowered in the least number of days. An increase or decrease in either day or night temperature from this combination increased the time necessary for flower development (7) . 43 The cultivars 'Polaris' and 'Bright Golden Anne' initiated flowers between 10° and 28° C under continuous light (24 hours a day). The fastest bud initiation occurred in the temperature range 16° to 22°. But only at the two lowest temperatures (10° and 16°) did the initiated buds develop into flowers (19). Chrysanthemums grown under so called split night temperatures (10 hours light at 22° followed by 4 hours dark at 16° and 10 hours dark at 10°, compared to 10 hours light at 22° and 14 hours dark at 16°) showed an average 4 days delay in flowering compared to 'normal' night temperature of 16°. The averaged temperature (for 24 hours) in this experiment with split night temperatures was 11.8° and at 'normal' temperatures 18.50 (53). A temperature regime with night temperatures of 17° for the first half of the dark period (8 hours); 10° for the remaining 8 hours and a day temperature at 22° resulted in a 3 day delay in time to flowering (52). Flowering was delayed when night temperatures were reduced for 7 1/2 hours from a constant 16° (8). As the low temperature duration (down to 10° C) increased from 6 to 10 l/2 hours during the night flowering delay increased from 4 to 11 days. lower ing night temperature from 16° to 13° C delayed flowering 3 to 8 days in experiments by Tsujita et a1. (66). The delay was due to retarded flower bud development rather than delayed flower initiation. Cathey (13) concluded that "the temperature during the dark period was 3.3 times more effective in hastening flowering than temperature during the light period" and that "the averaging of night 44 and day temperature or mean temperature was not correlated with flowering time". Cockshull et al.. (25) pointed out that the way Cathey (13) calculated mean temperature by averaging day and night temperatures without taking the number of'hours these temperatures were kept into consideration was misleading and made a recalculation of the result presented by Cathey (13). The new values showed that average temperature was important for flower development and that night temperature didn't have any special influence on time to flower by itself. Flower development in experiments conducted by Cockshull et a1. (25) was correlated with average temperature with time to bud appearance decreasing from 42 to 26 short days as temperature increased from 10° to 20°. Kohl and Thigpen (44) grew plants at night temperatures of 15.6° and 5.6°. Under the lower night temperature the development was 25 days slower. Just lowering the night temperature to 5.6° C for the first 21 days of short days resulted in flowering 6 days later than when grown at 15.6° C throughout. Night temperatures at _ 15.6° C for the first 3 weeks under short days and the remaining period at 5.6° caused a delay of 16 days (44). Zieslin and Kohl (72) achieved similar results with continuous night temperatures of 5.5° and l6.5°, however the delay in development was even larger (35 days) at 5.5°. Night temperatures lower than 16° during flower bud initiation increased the leaf number and stem length. The cultivar 'Shasta' initiated on average 10 more leaves at 2° night temperature than at 16° C and the stem length increased ca. 25 cm. Day temperature was kept at a minimum of 21° C (61). A.decrease in night temperature from 16° to 10° during the period from the start of short days until buds became visible gave an 45 increase in leaf number with 6 leaves (68). Mean internode length was longer under night temperatures of 15.6° than 10° C; the average length was 1.9 cm and 1.2 cm respectively. Day temperature was held at a minimum of 15.6°, but maximum temperature was uncontrolled and varied with time of year (69). When plants were transferred from 15.6° to 10° C, the third internode from the last leaf was expanding at time of temperature change and showed a significant reduction in length compared to when grown continuously at 15.6° C; the third and the sixth internode showed significantly increases in length when night temperature was changed from 10° to 15.6° C and the response was comparable to plants grown at 15.6° C throughout (69). Bonaminio and Larson (8) compared number of nodes on plants grown in an environment of 16° constant night temperature or when the temperature was reduced to 10° for part of the night. No difference in the number was found, but the plants under reduced night temperature grew taller. Internode length on plants grown in a climatic environment with different day and night temperatures was not the same as when a constant temperature corresponding to the average of the different day and night temperatures was used (25). At a lower day temperature the internode length was shorter and Cockshull et al. (25) stated "the main influence on internode extension was exerted by the day temperature, although this effect was accentuated in some cultivars if the following night temperature was low”. Height of plants at 21° - 23° day temperature and 10° night 46 temperature was found by Tawagen and Hassan (64) to be shorter than plants under 16° C night temperature. Bonaminio and Larson (7) reported similar results for Chrysanthemums grown under day/night temperatures of 30°/26° C and 18°/14° C. However, when Kohl and Mor (45) kept Chrysanthemums under a day temperature of 21°-27° C and night temperatures of 5° and 15.60 C, the plants under the lower night temperature grew taller. Several cultivars showed an increase in stem length when the temperature during part of the night was lowered from 15.5° to 10° C. This trend was even more accentuated when plants were provided supplemental light (ca. 40 umol s'lm‘z) during cloudy days. In this study day temperature was kept at 18° on cloudy days and at 22° on sunny days (52). The number of flowers per plant decreased with decreasing night temperature (from 27° to 10°), but the average flower diameter increased by ca. 1 cm from 3.8 to 4.7 cm (54). Cathey (13) found that high night temperature produced flowers with more petals. Chrysanthemums grown at 10° had 6 flowers per plant, while a temperature of 16° resulted in 12 flowers per plant (64). Several researchers (68,52,45) found an increased number of flowers at lower night temperatures. Increases in flower diameter with a decreased night temperature have been reported by Bonaminio and Larson (8) and Tsujita et al.. (66). The night temperature after visible bud appeared to be most important for flower size with an optimum temperature at 100 C. An even lower temperature at 4° C produced larger flowers, but the quality was "somewhat inferior" (68). For fast and high quality production of 47 Chrysanthemum Vince (68) recommended 16° night temperature up to visible bud followed by 10° - 13°. Increased leaf area under decreased temperatures has been found in several experiments (7,8,53). The rate of leaf emergence however, was faster as temperature increased (25). Several growth statistics (relative growth rate, unit leaf rate, leaf area, plant dry weight and leaf dry weight) were larger when the temperature was lowered for parts of the night from 'normal' night temperature (53). Ieaf Area Ratio and Leaf Weight Ratio decreased and Specific Leaf Area didn't show any difference between the higher and the lower night temperatures (53). Respitory dry weight losses in plants are temperature dependent. A decrease in temperature should decrease the respiration rate and reduce dry weight losses. The increased RGR supported this theory (45,53). Kohl and Thigpen (44) showed that the rate of dry weight gain could be the same if the Leaf Area Index was adjusted according to growth. The critical LAI in this experiment was found to be 2.7 - 3.0. If the LAI was kept at or above the critical value there was no difference in dry weight accumulation at 5.6° and 15.6° C. The efficiency with which the plants utilized provided Photosynthetically Active Radiation (PAR) was shown by Kohl and Mor (45) to be better at a night temperature of 5.6° rather than 15.6° C. This greater efficiency was explained to be due to a lower respiration during low night temperatures (45) . 48 Partitioning of Dry Matter The percentage dry matter partitioned to the leaves stayed the same (about 46%) when Chrysanthemums were grown vegetatively for 5 - 6 weeks at 20° C under different irradiances in the range from 1.9 to 9.2 MJ m‘2 d'1 (300 - 1470 umol s'lm‘z, 8 hr lightspan d"1). At a higher temperature (30°) the percentage partitioned to the leaves increased. The leaf area, however increased with lower irradiances and higher temperatures (1). Cockshull and Hughes (21) studied flower weight ratios (the weight of the flower divided by the total plant dry weight) in plants grown under different environments. They found that the heaviest plants always had the highest flower weight ratio. The proportion of the total dry matter going into the flowers was highly correlated with stage of flower development and the number of flowers per plant didn't seem to significantly influence the partitioning pattern. To improve flower weight and quality, either an overall increase in plant dry weight or a decrease in number of flowers per plant seemed to be necessary. Cockshull (20) and Cockshull and Hughes (21) have pointed out the importance of early disbudding to produce larger flowers. As the flowers developed, they increasingly became the primary sink for dry weight accumulation and the weight of vegetative parts became relatively constant. But when all flowers were removed dry matter was diverted into other parts of the plant, primarily the roots and to some extent the leaves. The extension of stems stopped when the flower buds were taken away, although the accumulation of dry matter continued in the stem and the weight per unit stem increased. From this 49 experiment Cockshull and Hughes (22) suggested that there was no severe depression in the rate of dry matter production per unit leaf area when the primary sink was taken away in Chrysanthemums. Woodson and Boodley (71) found that stems and petiols attained their maximum dry weight before the 8th week of growth when the fast flower development started in the cultivar 'Gt.#4 Indianapolis White'. The leaves however, continued to accumulate dry matter during flower develqament. The temperature in the greenhouse was kept at 24° day and 18° night, black cloth was used to provide a photoperiod of 15 hours dark and the plants were grown single stem. Under these conditions at least, the photosynthetic capacity seemed to exceed the demand from the sinks in this variety (71). LITERATURE CITED 10. ll. 12. 50 LITERATURE CITED Acock, B., D. A. Charles-Edwards and S. Sawyer. 1979. Growth response of a Chrysanthemum crop to the environment. III. Effects of radiation and temperature on dry matter partitioning and photosynthesis. Ann. Bot. 44:289-300. Anonymous. 1982. Floriculture crops. Production area and sales, 1980 and 1981. Crop reporting board, Econ. Stat., and Coop. Service. USDA, Washington, D.C. Anonymous. 1983. Yoder products for 1983-84 . . . something to grow on. Variety catalog for Chrysanthemums, carnations, geraniums, azaleas, snapdragons and foliage plants. Yoder Brothers, Inc. Austin, R. B., J. Bingham, R. D. Blackwell, L. T. Evans, M. A. Ford, C. L. Nbrgan and M. Taylor. 1980. Genetic improve- ments in winter wheat yields since 1900 and associated physio- logical changes. J. Agric. Sci. Camb. 94:675—689. Bertalanffy, L., von. 1957. Quantitative laws in metabolism and growth. Quart. Rev. Biol. 32:217-231. Blackman, V. H. 1919. The compounded interest law and plant growth. Ann. Bot. 33:353-360. Bonaminio, V. P. and R. A. Larson. 1978. Influence of potting media, temperature, and concentration of ancymidol on growth of Chgysanthemgm morifolium Ramat. J. Amer. Soc. Hort. Sci. 103: 752-756. and R. A. Larson. 1980. Influence of reduced night temperatures on growth and flowering of 'May Shoesmith' Chrysanthemums. J. Amer. Soc. Pbrt. Sci. 105:9-11. Briggs, G. E., F. Kidd and C. West. 1920a. A quantitative analysis of plant growth. Part I. Ann. Appl. Biol. 7:103-123. , F. Kidd and C. West. 1920b. A quantitative analysis of plant growth. Part II. Ann. Appl. Biol. 7:202-223. Carpenter, W. J. 1975. High intensity lighting of pot Chrysanthemum. Flor. Rev. 156 :19-20,59-60. Cathey, H. M. 1954a. Chrysanthemum temperature study. B. Thermal modifications of photoperiods previous to and after flower bud l3. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 51 initiation. Proc. Amer. Soc. Hort. Sci. 64:483-498. . 1954b. Chrysanthemum temperature study. C. The effect of night, day, and mean temperature upon the flowering of Ctlrysanthemum morifolium. Proc. Amer. Soc. Hort. Sci. 64: 499-502. . 1969. Chrysanthemum morifolium (Ramat.) Hemsl. p. 269-290. In L. T. Evans (ed.) The induction of flowering. Some case histories. Cornell Univ. Press, Ithaca. Causton, D. R. 1977. A biologist's mathematics. Edward Arnold, Ltd. London. and J. C. Venus. 1981. The biometry of plant growth. Edward Arnold Ltd. , London. Cockshull, K. E. 1974. Premature budding in year-round Chrysanthemums. Rep. Glasshouse Crops Res. Inst. 1974 :128-136. . 1976. Flower and leaf initiation by Chgysanthemum morifolium Ramat. in long days. J. Hort. Sci. 51:441-450. . 1979. Effects of irradiance and temperature on flowering of Chrysanthemum morifolium Ramat. in continuous light. Ann. Bot. 44:451-460. . 1982. Disbudding and its effect on dry matter distri- bution in Chrysanthemum morifolium. J. Hort. Sci. 57:205-207. and A. P. Hughes. 1967. Distribution of dry matter to flowers in Chrysanthemum morifolium. Nature 215:780-781. and A. P. Hughes. 1968. Accumulation of dry matter by Chgysapthemum morifolium after flower removal. Nature 217: 979-980. and A. P. Hughes. 1971. The effect of light intensity at different stages in flower initiation and development of Chgsanthemum mogifolium. Ann. Bot. 35:915-926. and A. P. Hughes. 1972. Flower formation in Chrysanthemum morifolium: the influence of light level. , D. W. Hand and F. A. Langton. 1981. The effect of day and night temperature on flower initiation and development in Chrysanthemum. Acta Horticulturae 125:101-110. Crater, G. D. 1980. Pot mums. p. 261-285. In R. A. Larson (ed.) Introduction to flor iculture. Academic Press, New York. Doorenbos, J. and A. M. Kofranek. 1953. Inflorescence initiation 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 52 and development in an early and late Chrysanthemum variety. Proc. Amer. Soc. Hort. Sci. 61:555-558. Draper, N. R. and H. Smith. 1981. Applied regression analysis, 2nd edition. John Wiley, New York. Erickson, R. O. 1976. Modeling of plant growth. Ann. Rev. Plant Physiol. 27:407—434. Evans, G. C. 1972. The quantitative analysis of plant growth. Blackwell Scientific Publications, New York. Fisher, R. A. 1921. Some remarks on the methods formulated in a recent article on "the quantitative analysis of plant growth". Furuta, T. 1954. Photoperiod and flowering of Chrysanthemum morifolium. Proc. Amer. Soc. Hort. Sci. 63:457—461. and K. S. Nelson. 1953. The effect of high night temperature on the development of Chrysanthemum flower buds. Proc. Amer. Soc. Hort. Sci. 61:548-550. Hughes, A. P. 1973a. A comparison of the effects of light intensity and duration on Chrysanthemum morifolium cv. 'Bright Golden Anne' in controlled environments. I. Growth analysis. Ann. Bot. 37:267-274. . 1973b. A comparison of the effects of light intensity and duration on Chrysanthemum morifolium cv. 'Br ight Golden Anne' in controlled environments. II. Ontogenetic changes in respiration. Ann. Bot. 37:275-286. and K. E. Cockshull. 1971a.' The effects of light intensity and carbon dioxide concentration on the growth of Chgsanthemum morifolium cv. 'Bright Golden Anne'. Ann. Bot. 35 :899-914. and K. E. Cockshull. 1971b. A comparison of the effects of diurnal variation in light intensity with constant light intensity on growth of Chrysanthemum morifolium cv. 'Bright Golden Anne'. Ann. Bot. 35:927-932. and K. E. Cockshull. 1971c. The variation in response to light intensity and carbon dioxide concentration shown by two cultivars of Chrysanthemum morifolium grown in controlled environments at two times of year. Ann. Bot. 35:933-945. Hughes, B. R. and M. J. Tsujita. 1981. The effect of supple- mental HPS lighting during propagation on rooting and quality of 'White Marble' cut Chrysanthemum. J. Amer. Soc. Hort. Sci. 106 :613-615. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 53 Hunt, R. 1978. Plant growth analysis. Studies in biology no. 96. Edward Arnold Ltd. , London. . 1979. Plant growth analysis: The rationale behind the use of the fitted mathematical function. Ann. Bot. 43:245-249. . 1982. Plant growth curves. The functional'approach to plant growth analysis. Edward Arnold Ltd. , Iondon. Jong J., de. 1978. Selection for wide temperature adaptation in Chrysanthemum morifolium (Ramat.) Hemsl. Neth. J. Agric. Sci. 26:110-118. Kohl, H. C., Jr. and S. P. Thigpen. 1979. Rate of dry weight gain of Chrysanthemum as a function of leaf area index and night temperature. J. Amer. Soc. Hort. Sci. 104:300-303. and Y. Mor. 1981. Producing pot Chrysanthemums at low night temperature. J. Amer. Soc. Hort. Sci. 106:89-91. Langhans, R. W. 1964. Light and photoperiod. p. 73-85. In R. W. Langhans (ed.) Chrysanthemums. A manual of the culture, diseases, insects and economics of Chrysanthemums. N. Y. St. Ext. Serv., and N. Y. St. Flower Grs. Ass. Inc. Machin, B. and N. Scopes. 1978. Chrysanthemums, year-round growing. Blandford Press Ltd. Poole, Dorset. Madden, L. V. 1980. Quantification of disease progression. Prot. Ecol. 2:159-176. Naggs, D. H. 1960. The stability of the growth pattern of young apple-trees under four levels of illumination. Ann. Bot. 96: 434-450. NbGree, K. J. 1981. Photosynthetically active radiation. p. 41-55. In 0. L. Lange (ed.) Encyclopedia of plant physiology, Vol. 12A. Springer-Verlag, New York. Milthorpe, F. L. and J. Moorby. 1979. An introduction to crop physiology, 2nd edition. Cambridge Univ. Press, New York. Parups, E. V. 1978. Chrysanthemum growth at cool night temperature. J. Amer. Soc. Hort. Sci. 103:839-842. and G. Butler. 1982. Cmparative growth of Chrysanthemums at different night temperatures. J. Amer. Soc. Hort. Sci. 107:600-604. Post, K. 1939. The relationship of temperature to flower bud formation in Chrysanthemums. Proc. Amer. Soc. Hort. Sci. 37: 1003-1006. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 54 . 1948. Daylength and flower bud development in Chrysanthemums. Proc. Amer. Soc. Hort. Sci. 51:590-592. and D. B. Lacey. 1951. High temperature produces long day effect on Chrysanthemums. Bull. N.Y. St. Flower Grs. 76:4. Potter, J. R. and J. W. Jones. 1977. leaf area partitioning as an important factor in growth. Plant Physiol. 59 :10-14. Richards, F. J. 1959. A flexible growth function for empirical use. J. m. %to 10:290-300. . 1969. The quantitative analysis of growth. p. 3-76. In F. C. Steward (ed.) Plant physiology: A treatise. VA. Analysis of growth: Behaviour of plants and their organs. Academic Press, London. Roberts, R. H. and B. E. Struckmeyer. 1938. The effects on temperature and some other environmental factors upon the photoperiodic responses of some of the higher plants. J. Agric. Res. 56:633. Samman, Y. and R. W. Langhans. 1960. Interactions of temperature and photoperiodism in Chrysanthemum morifolium. Proc. 15th Int. Ebrt. Congr. (1958) 2:400-411. Schwabe, W. W. 1952. Effects of temperature, day length and light intensity in the control of flowering in the Chrysan- themum. Proc. 13th Int. Hort. Congr. 2:952-960. Steel, R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics, 2nd edition. NbGraw-Hill Book Co., Inc. New York. Tawagen, A. M. and H. A. Hassan. 1974. Effect of temperature, daylength and cycocel on Chrysanthemum. Egypt. J. Hort. 1: 57—65. Thornley, J. H. M. 1976. Mathematical models in plant physio- logy I a quantitative approach to problems in plant and crop physiology. Academic Press Inc. , London. Tsujita, M. J., D. P. Ormrod and W. W. Craig. 1981. Soil heating and reduced night temperature effects on Chrysanthemums. can. Jo Plant. 801. 61:345-350. Venus, J. C. and D. R. Causton. 1979. Plant growth analysis: A re-examination of the methods of calculation of relative growth and net assimilation rates without using fitted functions. Ann. Bot. 43:633-638. Vince, D. 1960. Low temperature effects on the flowering of 69. 70. 71. 72. 55 Chrysanthemum morifolium Ramat. J. Hort. Sci. 35:161-175. and D. T. Mason. 1959. low temperature effects on internode extension in Chrysanthemum morifolium. J. Hort. Sci. 34 :199-209 . Williams, R. F. 1946. The physiology of plant growth with special reference to the concept of net assimilation rate. Ann. Woodson, W. R. and J. W. Boodley. 1983. Accumulation and parti- tioning of nitrogen and dry matter during the growth of Chrysanthemum. HortScience 18 :196-197. Zieslin, N. and H. C. Kohl. 1978. Effect of low night tempera- ture on 'Princess Anne' mums. Flor. Rev. 161:107-110. SECTION I INFUIENCEOFQUANHJMELUXDENSITYANDTEMPERATUREW FIOWERINGTIMEANDPIANT QWITY OF CHRYSANTHEMUM mRIFOLIUM RAMAT. 'BRIGHT GDLDEN ANNE'. Influence of Qlantum Flux Density and Temperature on Flowering Time and Plant Quality of Chrysanthemum morifolium Ramat. 'Bright Golden Anne'. M. G. Karlsson and R. D. Heinsl Department of Horticulture, Michigan State UniversityL East Lansing, MI 48824 Additional index words. central composite design, modeling Abstract. Chrysanthemum morifolium 'Bright Golden Anne' plants were grown under 15 combinations of Quantum Flux Density (QFD), day temperature, and night temperature in a central composite design. The influence of these environmental factors on flowering time and plant quality is reported both quantitatively and qualitatively. Time to flower depended on both irrad iance and the interaction between day and Received for publication . Michigan Agricultural Experiment Station No. . This project was supported in part by grants from the Fred C. Gloeckner Foundation, the American Florists Endowment, and the US Department of Agriculture Small Business Innovation Research program (grant no. 83SBIR-8-0001) in collaboration with Oglevee Computer Systems. Chrysanthemum cuttings were donated by Yoder Brothers, Inc., Barberton, OH. The cost of publishing this paper was defrayed in part by the payment of page charges. Under portal regulation, this paper therefore must be hereby marked advertisement solely to indicate this fact. lResearch Assistant and Associate Professor respectively. 56 57 night temperature. At a constant 20° C temperature, time to flower decreased from 90 to 60 days when QFD increased from 50 to 600 umol s‘lm‘z. Increasing either day or night temperature from 14° to 26° delayed flowering. High temperature delay was compensated for in part by increased QFD. Regression analysis showed shoot length to increaSe linearly as day temperature increased. Low night temperature accentuated the day temperature response. Total flower area increased as QFD increased or as night temperature decreased. Introduction While the influence of environmental factors on growth and development in many greenhouse crops has been extensively studied, few experiments have addressed several environmental factors simultaneously. Simultaneous evaluation of several environmental factors is important when determining the functional relationship between the environment and plant response. Commercially available computer systems for greenhouse climate control allow environmental control to be interactive. For example, temperature and C02 concentration can be controlled based on Quantum Flux Density (QFD) in the greenhouse (16). However to use this type of computer control system, one must know the functional relationship between environmental factors and subsequent growth and development of a particular plant. Both time to flower and plant quality are the primary factors of concern in commercial production of Chrysanthemums. While an extensive body of literature exists on Chgsanthemum moriifoliugg Ramat., we are unaware of any information describing the functional 58 relationships between Chrysanthemum growth and the environmental factors of day temperature, night temperature and QFD. This paper adresses this problem by describing such functions. Materials and mmods Rooted cuttings of Chrysanthemum morifolium 'Br ight Golden Anne' were planted individually in 10 cm pots and placed in growth chambers (Sherer-Gillete, Marshall, Michigan) under a QFD of 325 umol s‘lm'2 ( 16 hr d‘l) at a constant temperature of 20° C for seven days. On the seventh day after potting, short day (SD) photoperiod was initiated (10 hr light, 16 hr dark) and plants were pinched to six nodes and placed under appropriate treatment combinations (Table 1) with the thermoper iod following the photoper iod. Daminozide was applied 7 and 14 days after the start of SD at 2500 mg 1‘1. Ten days after the start of SD, lateral shoot number was reduced to 3 per plant. Lateral flower buds were removed when they had reached a stage where removal would not damage the apical flower bud. Shelves were lowered as necessary to maintain the desired QFD at the canopy top; QFD was measured with a Li-Cor LI-185B Meter and LI- l9OSB Quantum sensor. The QFD was provided by cool-white flourescent lamps (GE, F48T12, CW 1500) and incandescent lamps (GE, 40 W, 120 V) with an input wattage of 80:20 respectively. Average daily temperature fluctuated 11° C from the setpoint and QFD varied r. 10%. Plants were grown in a peat-lite medium (VSP, Michigan Peat Co.) and were automatically irrigated one to three times daily depending on plant size and environmental conditions using an 59 individual emitter in each pot. Nutrition consisted of 200 mg l‘1 N and K at every watering provided by ammonium nitrate, potassium nitrate and nitric acid (used to adjust water pH to 6.0). Necessary leaching occurred at each watering to prevent salt accumulation. A central composite statistical design (1,15,22) was used. Ranges for the three factors were 50 to 600 umol s"]-m"2 for QFD and 10 - 30° C for both day and night temperature (Table 1). Regression analysis was computed using the SPSS subprogram 'Regression' (25). Data were collected on five plants the day the plants were potted, at start of SD and every 10 days thereafter. Time to flower was determined as the day when half the flowers in the population had reflexed their outermost petals to a horizontal position. The QFD was measured at the canopy top and recorded when a plant was sampled. On each sample date, leaf area, leaf number, stem length, flower diameter and dry weight of these plant parts were collected on the original and three lateral shoots. Root dry weight was also determined for each plant at each sampling occasion. Only data on time to flower and final stem length, leaf number and flower size are reported in this paper. The remaining data will be published elsewhere (21). Treatments will be referred to with three numbers corresponding to QFD, day temperature, and night temperature, e.g. 50-20-20 is the first treatment in Table 1. Results and Discussion Time to flower depended on both irradiance and the interaction between day and night temperature (Table 2, Figure 1). A second order 60 equation to predict time to flower was developed (Table 3). The simple r values (Table 4) indicated QFD to be the main factor promoting early flowering. Simple r is the estimated first-order correlation between the dependent variable and the independent variable (25). At a constant 20° C day and night temperature, time to flower decreased from 90 to 60 days as QFD increased from 50 to 600 umol s‘lm‘2 (Table 2); at 325 umol s‘lm‘z, flowering occured in 70 days. The delay under 50 umol s"lm‘2 appears to be due to slowed development and not delayed initiation as flowering shoots in all three treatments had similar node numbers. This contrasts to work by others (10,11,12,18,27) which showed hastened flower initiation under high irradiance conditions. A possible explanation for this difference is that all plants in our experiment received one week of long days (LD) at 325 umol s'lm'2 (16 hr d‘l) prior to the start of the treatment environments. Since most plants can be grown successfully under 50 W m‘2 (ca. 250 umol s‘lm‘z) in growth chambers (2,9,28) 325 umol s"]-m"2 is a very acceptable QFD. In that research (10,11,12,18,27), plants were given the same QFD during both the LD and SD periods. Therefore sufficient carbohydrate reserves may have accumulated during the LD period in our experiment to allow rapid flower initiation even under the low QFD treatments while carbohydrate levels may have limited rapid flower initiation in the previously reported low irradiance treatments (10,11,12,18,27). The simple r values (Table 4) also indicated that both increasing day or night temperature within the experimental range delayed flowering. Treatment combinations with higher day than night temperature showed greater flowering delay than the reciprocal combinations (325-30-20 vs. 325-20-30 and 490-26-14 VS. 490-14-26). The 61 high temperature delay was compensated for in part by increased QFD. Under 160 umol s‘lm'“2 and a day and night temperature of 26°, flowering occurred after 90 days; an increase in the QFD to 490 umol s"]-m'2 decreased flowering time by 10 days. While high (above 30°) and low (4.5 - 10°) night temperatures were found to delay flowering initiation (14,26,30), Cockshull et a]. (13) reported the average daily temperature to be the factor controlling rate of plant development rather than the specific day or night temperature. The average temperature relationship did not hold in this experiment. Plants flowered in 70 days with average temperatures varying from 14° to 21° C (Table 2). Considering that shoots had similar node numbers (9-11 nodes) under a wide diversity of environments (Table 2), it appears no treatment environment specifically accelerated flower initiation but rather adverse environments delayed initiation. For example, plants in three treatments had significantly higher node numbers combined with delayed flowering; all three treatments are characterized by high temperature conditions (160-26-26, 325-30-20, and 490-26-26). High temperatures have been found to delay flower initiation and flowering in Chrysanthemum (8,30). The response to QFD and temperature explain why time to flower in Chrysanthemum does not vary significantly on a year around basis. In the winter, while greenhouse temperature control is good, irradiance limits the rate of development (Figure la). In the summer, irradiance is not limiting but greenhouse temperature regularly exceeds 20°. As day or night temperature exceeds 20°, flowering is delayed (Figure 1c). 62 Therefore time to flower in the Chrysanthemum on a year around basis is controlled by the relationship between greenhouse temperature and the QFD from solar radiation. Flowering time can only be minimized under conditions of high QFD and an acceptable temperature control. Irrespective of night temperature or QFD, plants grown in combinations with a high day temperature (above 20°) were tall with stem lengths about 30 cm. Plants grown at 14° day temperatures had an average stem length of 16 cm. Simple r values confirm these observations (Table 4). A plot of the regression equation predicts stem length to increase in a linear fashion as day temperature increases (Figure 2). Low night temperatures accentuate the day temperature response. Cockshull et a]. (13) also reported day temperature to be the main environmental factor controlling height in Chrysanthemum. They reported high day temperature caused stem elongation independent of what the night temperature was, although a low night temperature appeared to accentuate the effect of day temperature. While the number of leaves per shoot increased on plants grown under constant high temperature (26°), the average internode length did not increase compared to plants grown at 20° temperature. ()1 plants grown under high day and low night temperatures (160-26-14, 490-26-14 and 325-20-10), the opposite response occurred; no change in leaf number but average internode length increased (Table 2). Flower size is important for plant quality, but since people perceive the total flower area per plant rather than the diameter of individual flowers, total flower area was calculated. Simple r values (Table 4) indicate increasing QFD or decreasing night temperature to be the primary factors positively affecting flower area. As QFD increased 63 from 160 to 490 umol s"]-m"2 and night temperature decreased from 26° to 14°, total flower area increased from 34 cm2 to 304 cm2, almost a lO-fold increase (Table 2). The response to increasing QFD by itself or decreasing temperature by itself increased flower area, but to a smaller extent. The response to low night temperature has been reported before (4,29,30). Flower area was also reduced by high day temperature, especially in combinations with high night temperature (Figure 3). Part of the increased flower area per plant under higher QFD was due to increased uniformity of flower size on all three shoots (Table 2). While flower diameter varied from 1.2 to 7.4 cm on plants grown under 50-20-20, flower size only varied from 9.7 to 11.7 cm on plants grown under 600-20-20. Increased uniformity has been one of the primary advantages cited when plants received supplemental irradiation under low QFD conditions in greenhouses (11,23,24). In summary, increasing day temperature from 10° to 30° increased stem length, increased time to flower, and decreased flower area. Increasing night temperature from 10° to 30° slightly decreased stem length, increased time to flower, and greatly decreased flower area. Increasing QFD from 50 to 600 umol s"lm‘2 had no effect on stem length, but decreased time to flower, and increased flower area. High day temperature interacts with night temperature to further delay flowering and decrease flower area while a low night temperature interacts with high day temperature to increase stem length. While many of the influences of QFD and temperature on _C_h_gsanthemum morifolium flower time and plant quality have been previous reported, we believe this is the first time that all these 64 factors have been simultaneously reported both quantitatively and qualitatively. The real significance of this type of research lies in its ability to predict plant response under environmental conditions not specifically tested. For example, Table 5 shows the predicted time to flower, final stem length, and flower area per plant on plants grown under 4 different environmental conditions. The first two environments represent production under winter conditions in a northern UAR greenhouse with and without supplemental irradiation at 7511mol 571m.2 for 10 hr d"l (ca. 575 fc from high pressure sodium lamps). The last two environments represent summer production in a greenhouse at different temperatures. The supplemental irradiation during winter is predicted to decrease time to flower by 3 days, have no effect on shoot length, and increase flower area by 20%. Publications on the use of supplemental lighting have shown similar responses (5,6,7,l7,19). In the summer, decreasing day and night temperature is predicted to decrease time to flower by 10 days, decrease stem length 3 cm, and increase flower area by 90%. The direction and relative magnitude is expected (3,4,8,l4,20,29,30,31). The major limitation to this research is that plants were maintained under constant conditions throughout the development and this work represents one cultivar 'Bright Golden Anne'. Future work must be directed at determining the functional response to a changing environment, the vegetative growth period and to other cultivars. 65 o - o mm.H ON ON .oom mH H H H mm mm omw wH H H I H mm vH omv MH H I H H wH mm .omv NH H I H I H VH VH omv HH mm.H o o om ON mNm OH mm.HI o o OH ON mNm m o mw.H 0 ON om mmm m o mm.HI 0 ON OH mNm h o o 0 ON ON mNm m H H H I mm mm omH m H H I H I , mm wH owH w H I H H I wH 0N omH m H I H I H I vH VH omH N o o mm.HI ON ON cm H 92 . an GEO ucmHz woo ANIEHIm H081~ menace mmsHm> popoo . Aoov musumummEmB II omo ucmEummHB .gmHmmp ouHmomeoo Hmuucoo gnu cH pom: mcoHumcHnEoo ucoEummnu How mmsHm> oopoo ocm Hmsu04 .H oHnme 66 .2 ~85.» £338.. .3322... c 8 23!. awe-53.x. «5 3.839. a... 9.0.8.. 05 .6 a8 8 am co :3... 32¢ 8. 2 2. a... 2... a. 2 2 2 e2 2 2 8a 2 2. e2 .6 a... 2.2 2 .n 8 22 2 2 8.. 2 n... em as .6 t. 2. 2 2 e2 2 z 2.. .3. .2 no a... e... .2 2 .n 2 22 z 2 2.. 2. 3... 2.2 a... a... t. .. 2 2 a... z z 8. 2 s .3 .6 in 2.. .. 2 8 22 2 2 2n 2. 2 2. a... .2. .5 2 2 2 N... 2 2 2m 2 2 2.. e6 3 e2 : 2 2 «.2 2 2 2.” 3 2. a... a... .6 2. a 2 2 2.... 2 2 2» «2 a.» a... a... 2.2 2 2 2 22 2 2 2n 8. .2 92 o... 2.2. 22 .. 2 2 e2 2 2 2.” a2 .22 «.2 5.. .2 2 2 2 22 2 2 2n 2. 2 a... a... e; 9.. 2 2 8 22 2 2 8. 2 a .2 .... e... 2.. .. 2 2 e2 2 z 8. 2 .2 .3 a... .2 .5 .. .n 8 92 z 2 2. 8 2. a... .5 a... 2. 2 2 2 a... z z 8. 2 8. 2.. as I 2.. .. 2 2 e2 2 2 2 hm”... “Hanan“ Lawns: n .85. 2 ~85 . 32.» 25W. 83”....“ 7”. 59%. £2". Soweto 2...... a... c... -38... a 3 S... .2 t .5 c :6 .Ifll ILIII 2.3.. .33 .8. 5332.. no.8... 89:3... .3. a: 3 an.» 8982 no ... on... 28.3.835 .052 .828 23.5. 328...... llofieamxsfi so accuse—2.3 a..- 5393 .3 9.39.0.qu as... 9..- 832093 a3 .93 so 3531—... .~ 03-» 67 assuasmanu aza.z u h: .mgausroasou ago.» he .xu.m=ma x=.u saucsac u ouQN .... .-2ee.. - .-2222. .2 x .2 .-22e2. - 2-22am. 2-222.. - .z x are 2.22.2. 2.22.2. 2-22... - .2 x are .22. - .-2eem. .-22.m. 2..z. 22..». 2-222.. - .-22.a. 2..a. 2-2222. - 2-2222. - 2.2222. 2.2.2. 2.2.2. 22.2- as..- .2 2.2.22 22.. 222.2- .2 222. ..222.. .-2.22. - ale em..22e- ..e.2. e.e.2.. .ee..ee2 a... manners... swans“... 2%....“ 2...; Manama . oes< cospow 2:m..m. e=.—ow.soe 2:36gusangeo c. ace—a sea no.2 swche use games. uooem .gozo.m o» as.» so» muco.o.$emou sopmmosmma .m apnea 68 Table 4. Simple r values for time to flower, shoot length and flower area per plant in Chnysanthemum morifolium 'Bright Golden Anne'. Z Regression Simple r coefficient Time to flower §hoot length Flower area QFD -.58 -.01 .57 OT .53 .87 -.17 NT .35 -.19 -.62 (QFD)2 -.55 -.04 .57 (or)2 .57 .86 -.21 (NT)2 .36 -.15 -.63 QFD x DT -.24 .45 ..40 QFD x NT -.34 -.10 .13 OT x NT .63 .47 -.55 zQFD = Quantum Flux Density; DT = Day temperature; NT Night temperature 2.222 2.22 2... ,_ 2. 22 222 2.22. .22 2. .2 22 22 222 2.2.2 2.22 ...2 2. 22 222 2.22. 2.22 2.22 2. 22 2.. 9 6 Auguv Asov A2222. .mzo—w 2:222 22: .m-spumpoanv 222 .222: 5222. 2822 3 2...: 322 222. 2.2 vouopum.2 acmsco._>=m .mucmsco._>=m “22.22.22 .22. .222: 232.2 .m:=< 222.2w 2222.2. 52.—2.2.22 5252;222mxggo .2. um»o.2m.2 22.2 .232.. 2:2 guoem. 22°22 .22.. ..232.. ca 252. .m 2.22. 70 Figure 1. Predicted time to flower as effected by day tenperature, night tenperature, and QFD for Cngysanthenum norifolium 'Bright Golden Anne', at a) 100 11ml 3"1m"2 ; b) 250 umol s‘lm‘z; and c) 400 mm s'lm“ . TIHE T0 FLOHER (DRYS) TIHE TO FLOHER (OHYSl (DRYS) TIME TO FLOWER (FD 100 and 8“!!!" mm Ylnnnnfuu .- -0- 12° 6 .. -abao° c .. ¢ ’ 41-200 c 10 :l‘ :1. #22 423 7L 30 any TEHPERRTURE (CENTIGRRUE) L g L QFD 250 and s"m'2 MG"? TEMPERAIURE -0- 13° c -£P20¢ C -n-2l° c .. ‘----:‘: ; b 3' ; ¢ ‘ 4* ‘ 4% : ¢ ¢ 10 t4 to 22 28 so DRY TEHPERRTURE (CENTXGRRDEl QFD 400 pmol s"m'2 to j 0 NIGHT unreal-Tune -o- 12° c 95” +200 c -N-28° c /X 85-0 / )1D " 4D o . . V r V r r r 10 14 to 22 '447 :20 so DRY TEHPERRTURE (CENTIGRHDE) Figure 2 . 72 Predicted final shoot length as effected by day tenperature and night tenperature for Chrlsanthemum morifolium 'Bright Golden Anne' at 325 umol s‘lm‘z. om Hmocmoszwo. mmapcmumzm» >¢o @N NN flu ta OH -# v w #- v u w u w . u m 6N .. o 222* ..\u_ 0 222-2.. .22 .\ 0 our :0: m¢=P80 ANrEHIm HOEHV Hones: mmsHm> pmpoo Auov wusumummflma omo ucmsummua .cmHmmp muHmOQEoo chucmo on» :H poms mcoHumcHQEoo ucwEummHu Hem mmSHm> empoo new Hmsuog .H mHnma «use uses» .9 Hanan use less nave on» ac-sgxo.» Hon o..I m.. a~. a..I mu. .n. ~.. ~—. me. an. o—.I p: x H: «H. _..I- .9. on. .I us. as. us. as. Hm. as. p: x ago “a. Hm.I u.. n:.. .~.I «a. go. No. as. Ha. mm. p: x as: m~.I an. ~..I ~o.I on. up. mo.I we. on. .o. no. N p: on.I Hu. 0.. no.I um. an. o.. o—. mm. .m. on.I ha a. 3.- S. a. .I 8. a. 3. 3. 3. 3. ~35 -.I on. a..I «a. an. H». me.I ac. an. co. me. p: Hn.I -. um. mo.I an. 9.. cm. n—. an. .m. .I H: an. n~.I H.. an. an. no. no. so. so. so. no. ama .oa—-» 5 one... acoHu—beoou soHanognog au .~ mu - o. n.»- a... ... o.n ~.n o.n on cw coo - an us a on e..— n.~— 9.. o.m H.. m.—. cu am am. ~n nu a~ ~H oh a... m.o o.n N.» ..n ..p ow .— ea. Hm an an a as n.._ s..— m.. a.» n.u n.— .— on co. Hm cu on n. as a... ..—— 5.. a.» ..n m.— .— ._ co. nu Ha an .n 0 ea ..o— ..s o.~ n.n —.n u.e an ca m~n a. mu mm c. a as o..- ..e p.» a.~ o.. 9.. o. as m~n Nu um .. m an ..—. .m.u m.~ ..n m.. m.: ow on awn - m~ mu nu a. 5.. a.» o.. ~._ ~._ m.— =~ a. m~n an 8 .u 2 2 Q6 n6 .6 ad ..~ H; 8 8 an an aw m~ Hp as ...H ~.a ... a.~ o.~ n.— ca ea mun n~ an «n s co ..a ..o ~.~ o.» .o.n e.g cw cu on. mm H» on m. as ..s 5.. 9.. ~.~ ~.~ —.. ow, .— oo— HN .n on e on 9.. a.» o.- ..N m.~ ..o ._ o~ oo— a~ an o~ n. as ¢.m n.» o.. o.— m.— e.g .— .— co. u~ .. _~ a so ... ..~ p.. o.— e.g ..e on aw em usage—u uo>aoa any“. «aces gaze—u o:.soxo.e so again on a. «gaze—u nos-«H u-oum «Hoax » .«L- Iuwalav a» u p 9 ~23... a... .82.... £3 3 2.2... .5 2.... .38 .3 ~22... I. .. 2.1.8.2.. .2... .328 2.2.... 3:228 1555?... 5 2.2.2.3.... e... 8.3.58. 2...... a... .2: .3 .2335... as... 2. 3. .2: .3 .88....— .~ .3: 91 mta.armasm. .53.: I .z .m.=...maso. Ha: I .a ..H.I=ma x=.. saueuao I sac. .Imam.. .Immmm. I .Imm.m. .Imma.. - ~-m~.m. I .z x .a .Imm.m. ”Ian... I mIm~.a. m-..... I m...... I .z x ale «Ian... «-mm... - H...@.. I ”-meam. MImamm. .a x ale .Imwm.. I .Ime.m. I .Imwc.. .... .Imoae. I ~..z. ~I..... I .e.. I NIHNHH. .Im.me. NIma... ~..a. m-m.m.. m...... I .IHH... .I..... I «-me.n. ~.a.a. mom. I cmc.. mam..- waN..- we... .2 a... I em... Nae..- ~.m.~- .e.... .a HIHamm. ac.. .Imcoa. I NImmma. .Imemw. I sic can... ma..so- e....s .w~.mm .cm... H=I.I=ou .N..I .. .ow. I .. .am. I .. .N.. I t. .m.. I .. .=I.u...mou Hempmzwxgu mzapm: ngu Hga.m:~».u .ga.m3~».u asm.mz~x.u ~=opmmmgmmz —muou pmcpm .m3opu Hemutma mom. «smegma emu. «smegma ace. acmugma .m:=< =mu.ow ago—gm. 3:..8...os 225mguemmxtgu :. usm.m3 Hg. .auo. .m:.. can mega.m3 syn .mzo.. can .umm. .emum .Hoo. Hemutma .8. mucm.u..mmcu co.mmm.mm¢. .m mHamh 92 ..N. m.~m ...~ o..m m.~. m. .N cc. o... m..~ m.- m.om m.o. mm mm cc. ..m «.mm ..m~ m.c~ m... o. om omm o.m H.om m..m m.mm m... m. cm m.— E E E E E 2.... E. .N-.. -22.... Hem.mz an. Hem.m3 Hum Hem.mz syn H;m_mz amp msmpmz age Hum. asmp omo Hem—a .muow .mzo.. mamH amum woo: _ um.u.vmga Heme=o..>:u .m::< emu—om Hem..m. s:..a...os snemguemmaggu so. mgmzo.w use mm>mm. .mEmum .muoo. o. mmco.mpm.aa meo.agoao.a new Hem.mz Hg. .auou .ocp. umuu.umga .. m.am» 93 Figure 1. Predicted final plant dry weight as effected by day tenperature and QFD for Chrysanthemm morifolium 'Bright Golden Anne' at 200 night temperature. om Hoe. mmDH¢mmmZMH >¢o mm mm mHI .. o. 1b -II- .1- «(I- r p 1 JI d p p d 4 NIETm .08.. OOVIXI «IE?» .055 OmNImI «IE?» .08: co. 1? 9.0 L .m. DION mm:.¢mmazm. .Io.z [0) lH‘OIEIM A80 iNU‘IcJ "IUNIJ 95 Figure 2. Predicted proportion of dry weight partitioned to the roots as effected by day tenperature and QFD for Qirysanthenum morifolium 'Br ight Golden Anne' at 20° night temperature. Hoa. mmawmmmmzm. >¢o on on «N m.. .H ‘P o. P d .- HI- fir- qu- 8 4b P 4 NI EFI «IE?» 3:... co. Ix. a .05: 6mm Iml «IE?» .08.. co. I? . Duo DION mm:.¢mmazm. .Io.z [%J iHflISM A80 1008 97 Figure 3. Predicted prqaortion of dry weight partitioned to the stems as effected by day tenperature and QFD for Chrysanthemum morifolium 'Bright Golden Anne' at 200 night tenperature. Ho... umapmmmmzmp Ea 8 mm mm 2. I o. P Ip P 1 d d .1!- P G 1 HI- mIETm 3:... 00.1? «IE—Ia. _OE= OmN Iml to. «IETm .08: co. .0I ouO 0 com mmapmmmmzm: .1..on (96] iHOIEM A80 14318 99 Figure 4. Predicted proportion of dry weight partitioned to the leaves as effected by day tenperature and QFD for Chrysanthemum morifolium 'Bright Golden Anne' at 20° night tertperature. Hue. mm:.¢mmmzmH »¢o om mm mm m.. .. o. h F b 8 n p d u H a J d O “F p 8 d «IE?» 2...... 2...... «IE...» .053 omNImI «IE...» .08.. oopIeI .umnv 1 6w Doom manpcmmmzmh Hron [96] iHOISM A80 383'] _ 101 Figure 5. Predicted proportion of dry weight partitioned to the flowers as effected by day temperature and QFD for Chgysapthemim mgifolium 'Br ight Golden Anne ' at 200 night tenperature. on .0... mmahmmmmzm. Ea mm mm 8. 2 o. P qr- P P b 8 O 1 1 T J 1 d J «IETm .02... 02.1.? NIEFIa .OE... OmNIEI «IE?» .08.. co. I0I Duo l .0. Q cow mmzpqmmnEMC 5.6.2 leJIBM A80 831401;! (96] 103 Acock, B., Charles-Edwards, D. A., and Sawyer, S. (1979). Growth response to the environment. III, Effects of radiation and temperature on dry matter partitioning and photosynthesis. Ann. Bot. 44,289-300. Armitage, A. M., Carlson, W. h., and Cress, C. E. (1981). Deter- mination of flowering time and vegetative habit of Tyetes gtula through response surface techniques. J. Amer. Soc. hort. Sci., 106,632-8. Bj6rkman, O. (1981) . Responses to different quantum flux densities. p. 57-107. In 0. L. Lange (ed.) Encyclopedia of plant physio- logy, Vol. 12A. Springer-Verlag, New York. Blackman, G. E., and Black, J. N. (1959). Physiological and eco- logical studies in the analysis of plant environment. Ann. Bot., 23,51-63. Bula, R. J., mykerd, C. L., and Langston, R. C. (1959). Growth response of alfalfa seedlings under various light regimes. Agron. J., 51,84-6. - . Carpenter, W. J. (1975) . High intensity lighting of pot Chrysan- thm. Flor. kVo' 156,19-20'59-600 Carpenter, W. J. (1976) . Photosynthetic supplementary lighting of spray pompon Chgysanthemum mogifolium Ramat. J. Amer. Soc. hort. Sci., 101,155-8. Cockshull, K. E. (1982) . Disbudding and its effect on dry matter distribution in Chrysanthemum morifolium. J. hort. Sci., 57, 205-7. Cockshull, K. E., and Hughes, A. P. (1967). Distribution of dry matter to flowers in Chrysanthemum mpifolimn. Nature, 215, 780-1. Evans, G. D., and Hughes, A. P. (1961). Plant growth and the aerial environment. I. Effects of artificial shading on Lugtiens miflora. New Phytol., 60,150-180. Gardiner, D. A., Cragle, R. G., and Chandler, P. T. (1967). The response surface method as a biological research tool. Tenn. Agric. Expt. Sta. Bul., 429,25-40. Hammer, P. A., and Langhans, R. W. (1976). Growth models for Helianthus annuus L. and Zinnia elegans Jacq. J. Amer. Soc. 104 horto $10, 101,475-90 Hicklenton, P. R., and McRae, K. B. (1984). Vegetative growth and flowering of pot Chrysanthemums in response to supplemental radiation and split night temperatures. J. Amer. Soc. hort. Sci., 109,30-3. Hughes, A. P., and Cockshull, K. E. (1969). Effects of carbon dioxide concentration on the growth of Callistephus chinensis cultivar Johannistag. Ann. Bot., 33,351-365. Hughes, A. P., and Cockshull, K. E. (1971). The effect of light intensity and carbon dioxide concentration on the growth of Chrysanthemum morifolium cv. Bright Golden Anne. Ann. Bot., 35,899-914. Hughes, B. R., and Tsujita, M. J. (1981). The effect of supplemental I-IPS lighting during propagation on rooting and quality of 'White Marble' cut Chrysanthemum. J. Amer. Soc. hort. Sci. 106,613-15. Karlsson, M. G., and Heins, R. D. (1984). Influence of quantum flux density and temperature on flowering time and plant quality of Chrysanthemum mpifolimm Ramat. 'Bright Golden Anne'. J. Amer. Soc. hort. Sci. (in review). Karlsson, M. G., Heins, R. D.,_ and Carlson, W. H. (1983). Devel- opment of environmental strategies based on plant growth models. Acta Horticulturae, 147, (in press). Kohl, H. C. Jr., and Mor, Y. (1981). Producing pot Chrysanthemums at low night temperature. J. Amer. Soc. hort. Sci., 106,89-91. Merritt, R. H., and Kohl, H. C. Jr. (1983). Crop productivity efficiency of petunias in the greenhouse. J. Amer. Soc. hort. Sci., 108,544-8. Nie, N. E., Hull, C. H., Jenkins, J. G., Steinbrenner, K., and Bent, D. H. (1975) . Statistical package for the social sciences (SPSS) , 2nd edition. NbGraw—Hill Inc., New York. Parups, E. V., and Butlar, G. (1982). Comparative growth of chry- santhemums at different night temperatures. J. Amer. Soc. hort. Sci., 107,600-4. mykerd, C. L., Langston, R., and Mott, G. O. (1960). Effect of intensity and quantity of light on the growth of alfalfa, red clover, and birdsfoot trefoil. Agron. J., 52,115-9. “11111111111111 “WIWISIWTIIYI 111111111158 3 1293 03083 1378