Growth Differences Within Full-Sib Families of Picea Abies (L.) Karst

STUDIA FORESTALIA SUECICA

Growth differences within full-sib families of Picea abies (L.) Karst.

Tillvtixtskillnader inom helsyskonfamiljer av gran Picea abies (L.) Karst

GOSTA ERIKSSON Department of Forest Genetics, Royal College of Forestry, Stockholm, Sweden

VELICHKO GAGOV Higher Forestry Institute, Chair of Forestry, Sofia, Bulgaria

SKOGSHOGSKOLAN ROYAL COLLEGE OF FORESTRY

STOCKHOLM Abstract

ODC 165.3: 174.7 In the present paper the reasons for differences in plant height within full-sib families of Norway spruce are discussed. In many cases it was shown that the duration of the growth period to some extent was responsible for the differences in plant height. If the total plant height is plotted against the percentage growth 4 or 5 weeks before growth termination, clones with good growth rhythm characteristics could conveniently be identified. Therefore this technique is recommended for selection of ortets in nurseries.

Ms received 4th October, 1975

LiberForlag/Allmanna Forlaget ISBN 91-38-02753-4, ISSN 0039-3150

Berlingska Boktryckeriet, Lund 1976 Contents

1 Introduction ...... 5 3.3 Total plant height-total branch length ...... 20 2 Material and methods ...... 6 4 Acknowledgements ...... 22 3 Results and discussion ...... 8 5 Summary 23 3.1 The point of time for 90 percentage ...... growth ...... 8 6 Sammanfattning ...... 24 3.2 Growth at certain intervals before growth cessation ...... 14 Literature ...... 25

2 .SFS nr 130 1 Introduction

The different growth capacity of prove- terminated their growth later than the nances of Picea abies originating from others. varying countries has long been known It is also evident that there are dif- (e.g. Langlet, 1964, Gellander, 1970, ferences in growth capacity within full-sib ICrutzsch, 1975). These dissimilarities be- families. Therefore, the question might be tween provenances can, at least to a certain raised as to whether these differences can extent, be attributed to different photo- also be attributed to varying photoperiodic periodic responses of the provenances responses of the individual plants or whether tested (cf. Sylven, 1940, Dormling, 1973). they are due to some other physiological Similar results for provenances of Picea reasons, such as photosynthetic capacity. sitchensis (Burley, 1966) and Pinus taedu 'The purpose of the present investigation (Perry et al. 1966) as well as for clones of was to throw some light on the question Salix alba/fragilis (Lattke, 1973) were re- "Why do some plants from a particular ported. This means that budset and growth full-sib family grow better than their full- cessation take place following exposure sibs?" The benefits derived from the present to dark periods of different length in dif- investigation might be twofold: ferent provenances. Such differences are 1. To guide the physiologists in under- not limited to the provenance level. Thus taking more detailed investigations con- Hagner (1970) demonstrated a great varia- cerning the reasons for the variation of tion within provenances of Pinus sylvestris growth capacity within a full-sib family. as regards the annual rhythm. Similar ob- servations have been noted by Schiitt (1962) 2. To guide the geneticists in their work for Pinus sylvestris and by Kozlowski and of selecting ortets which should be used Peterson (1962) for Pinus resinosa. The as starting material for commercial pro- tallest trees became tallest because they duction of cuttings. 2 Material and methods

In an investigation studying the inheritance The short CN was represented by clones of the critical night length for budset (CN) from northern Sweden and the long CN in Picea abies (Dormling et al. 1974) the was represented by clones from France. plant height was measured once a week. Further details as regards the clones are These data were used in the present in- presented below (Table 1) and in a paper vestigation. Three types of full-sib families by Dormling et al. (1974). All plants, were included in the study: irrespective of type of the progeny, were grown under continuous light for 11 weeks. short CN x short CN From then on the plants were grown under short CN x long CN different light regimes (0-8 hours of long CN x long CN

Table 1. The correlation and regression coefficients for the relationship between the date for reaching 90 per cent of the total plant height and the total plant height.

Number of Exp. r dark hours No.

French x French

Swedish x Swedish

French x Swedish

9 0 1006 x AC 2003 7 10 0 1007 x AC 2003 7 10 0 1007 x AC 2003 6 11 0 1008 x AC 1008 7 12 Z 3008 x Neufchateaux 7 20 0 1006 x BD 2006 6 2 1 0 1007 x BD 2006 6 darkness) for six weeks. So far six progenies be mentioned that each progeny was repre- from each type of combination have been sented by 16 plants growing in four-plant studied. The experiment was partitioned in plots. The two French x French combina- such a way that 12 of the progenies were tions were represented by fewer plants in studied at one time whereas the other six the second experiment. To get reliable in- were tested in a later experiment in which formation on the response of plants to a two progenies from the first experiment certain photoperiod it was necessary to were included a second time, thus con- study the progenies which at the termina- stituting a reference between the two ex- tion of the experiment showed 100 per cent periments (cf. Table 1). Moreover, it may growth cessation.

3 - SFS nr 130 3 Results and discussion

In Figure 1 plants belonging to progeny 8 The correlation and regression coeffi- (Swedish x Swedish) are illustrated. The two cients for the relationship between the point most extreme plants are shown, as well as of time for reaching 90 per cent growth some intermediate plants. It is obvious from and the total plant height were calculated. this photograph that there is a great varia- The coefficients are listed in Table 1. From tion within a full-sib family as regards the this table it may be seen that the correlation plant height. coefficients 18 times out of 20 were posi- Before entering upon the discussion of tive and nine of these cases were significant. the growth pattern within full-sib families, The progenies which show a negative cor- the general pattern of growth in the three relation are all weak. However, they are categories of progeny will be commented. In interesting from a breeding point of view as Figures 2-3 the percentage of growth of will be discussed below. The large dif- the individual progenies are illustrated. ferences as regards the correlation and From these figures it may be seen that there regression coefficients from progeny 10 in is a slight difference between the Frenchx the two experiments may partly be ex- French progenies on the one hand and the plained by the difference in photoperiodic French x Swedish and the Swedish x Swedish conditions. These plants may respond faster ones on the other hand. Thus none of the to a longer dark period than to a shorter. French x French progenies had reached their Anyhow, this observation indicates the 90 percentage growth four weeks before the necessity of testing the plant response under termination of the experiment, whereas all outdoor conditions. The regression coeffi- French x Swedish and Swedish x Swedish cients indicate that the duration of growth progenies reached their 90 percentage is to some extent responsible for the ob- growth before that point of time. This served difference in plant height. The observation is promising since it indicates longer the growing period the taller the the possibility of using those particular plant. This in turn means that there is a provenance hybrids in practical forestry. variation of the photoperiodic response even within a full-sib family. For each type of progeny the two ex- 3.1 The point of time for 90 percentage tremes as regards the regression coefficient growth were selected for an illustration of the To test whether or not the variation in final relationships discussed above. Progenies 16 plant height within a full-sib family could -17 contained fewer plants as pointed out be attributed to differences in the duration above and were therefore excluded from of the growing period, the final plant height the diagrammatic illustrations. Figures 4-9 was plotted against the point of time for reveal that there is a great variation within reaching of 90 percentage of the total each type of progeny. The strength of the growth (cf. Figures 4-9). This method will photoperiodic response seems to be less be referred to as method I. This percentage pronounced in the French x French pro- was preferred to the date for complete genies. cessation of growth. At the 90 per cent These diagrams should be used as a level the growth curves have not yet reached starting point for a discussion of the selec- their insensitive parts where they start to tion of ortets for cutting propagation. It is flatten. probable that reforestation in future would Figure 1. Photograph of plants belonging to progeny 8, AC 1002xAC 1009. The biggest and smallest plants are illustrated as well as some intermediate plants.

ERCENTAGE OF GROWTH

WEEKS BEFORE TERMINATION OF THE EXPERIMENT Figure 2. The relationship between the percentage growth and time for different types of progenies in experiment 1. F = French; S = Swedish. PERCENTAGE OF GROWTH

WEEKS BEFORE TERMINATION OF THE EXPERIMENT Figure 3. The relationship between the percentage growth and time for different types of progenies in experiment 2. F= French; S=Swedish. The dotted line refers to progeny 18 which had to be studied at 6 hours of darkness since 4.5 hours of darkness did not cause a 100 percentage growth cessation. gain from using cuttings. This is particularly fore be regretted. In progeny 2 (Figure 5) pronounced for Norway spruce (cf. Klein- one plant is marked. This plant possesses schmit 1974). To avoid late spring and early a good growth capacity and terminates its autumn frost damages in Norway spruce growth comparatively early. Such plants plants it is necessary that the plants show should be searched for in the selection of an annual rythm that avoids the risks for ortets for cutting propagation. such damages. Therefore, the bud flushing The growth curves for two extreme plants should not take place too early, whereas the from progeny 10 are illustrated in Figure bud set and growth cessation as well as the 10. lignification should not take place too late The variation in photoperiodic response (cf.-e.g. Krutzsch 1975). Therefore, it as studied in the present investigation is seems advantageous to select ortets among summarised in Figure 11. This figure reveals plants which show a good growth capacity that the photoperiodic response is on an although they terminate their growth early. average considerable, amounting to ten or If the plants in progeny 10 are investigated, more days. The absolute minimum of 4.4 hardly any plant would fulfil this criterium days was noted for progeny 12, while the for selection. The situation is reverse among maximum was 22.6 (progeny 9). This great the plants of progeny 12, in which the tallest variation is promising for the forest tree plant was also the first one to reach 90 breeder but of less advantage for the percentage growth (cf. Figure 9). The over- forester in his search for a hardy material all poor growth of progeny 12 must there- showing limited variation. Figure 4. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 1 (French x French).

Total plant height, rn

Figure 5. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 2 (French XFrench). Total plant height, m ~lllllllllllllllllllllllllllllllllllllll111lll11I1~11l11111l1l1lll1111l1lll~lll#1111ll1lll11111lHll1~ I I 0m34 I 0 1007 x AC 2003 I I I I 1 I I I I I I I I I I I I I I I I I I I -I I I I I I I I I . I I I I I I I I I I I I I I I I I I I I I I I I I I I I 0.2 I I I I I I I I I I I I I I I I I I I I I I I I I . I I I I -I I I I I / I I I I I I I I t I I I I I I I I I I I I . I /* :: 0.1 2 I I I I I I I I :II~~R~~~III~IIIIII~IIIII~IIIIIIIIIIIIIIIIII~I~IIIIIIIIIIIIIII~IIIIIIIIIIIIIIIII~IIIIIIIIIIIIIIIII~Z 11.5 12 13 14 Point of time -weeks- for reaching 90 percentage growth Figure 6. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 10 (French x Swedish).

Figure 7. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 12 (French x Swedish). Total plant height, m

Figure 8. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 7 (Swedish x Swedish).

Figure 9. The relationship between the total plant height and the date for reaching 90 per- centage growth of individual plants of progeny 8 (Swedish x Swedish). Time, weeks Figure 10. The growth pattern of the two most extreme plants in progeny 10 (French x Swedish).

Whether or not this great variation is measurements once a week during the due to the cultivation technique-1 1 weeks growing season. Therefore, it cannot be of continuous light and then six weeks of applied in investigations comprising large 5-7 hours of darkness per day-must be materials. To connect the method used tested in separate experiments. If this cul- above and one which might be possible to tivation technique brings about a greater handle on a large scale, the photoperiodic diversity of the material it means that the response was also studied by relating the selection would be facilitated. total plant height to the percentage growth at certain time intervals before the termina- tion of the experiment (referred to as 3.2 Growth at certain intervals before method I1 below). From Figures 2-3 it growth cessation may be seen that 4-5 weeks before the The way of studying the photoperiodic termination of the experiment most closely response of plants as presented above is, of corresponds to the 90 percentage growth course, a laborious way, with the regular level. Therefore, the relationship between Table 2. The correlation and regression coefficients for the relationship between total plant height and the percentage growth six, five, and four weeks before the termination of experiment 1.

Number of weeks before the termination of the experiment 6 5 4 Progeny Numbers No. of dark hours r b r b r b

Swedish x Swedish 5 6 7 8

French x Swedish 9 10 11 12

French x French 1 2 3 4

Table 3. The correlation and regression coefficients for the relationship between total plant height and the percentage growth six, five, and four weeks before the termination of experiment 2.

Number of weeks before the termination of the experiment 6 5 4 Progeny Numbers No. of dark hours r b r b r b

Swedish x Swedish 6 18 19

French x Swedish 10 20 2 1

French x French 16 17 Davs

Figure 11. The average difference in days be- tween the first and the last plant within a full-sib family reaching 90 per- centage growth. The av- French French Swedish erages refer to the three X X X types of progenies stu- French Swedish Swedish died.

the total growth and percentage growth of Figures 4-9 and 12-17 reveals that four or five weeks before the termination they more or less constitute a reflected of the experiment was studied. In addition, image of each other. The strong agreement the data from six weeks before the end of between the results of the two methods the experiment were used. The same pro- is further supported by an analysis of the genies that were shown in Figures 4-9 are relationship between the correlation and also illustrated by making use of method I1 regression coefficients of the two methods (cf. Figures 12-17). A pairwise comparison as shown below:

Relationship Comparison between: correlation coeff. regression coeff. method I-method I1 4 weeks method I-method I1 5 weeks Final plant height, m

Percentage growth 35 days before the termination of the experiment Figure 12. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 1 (French ~French).

Final plant height, m

Percentage growth 35 days before the termination of the experiment Figure 13. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 2 (Frenchx French). Final plant height, m

Percentage growth 35 days before the termination of the experiment Figure 14. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 10 (French x Swedish).

Figure 15. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 12 (Frenchx Swedish). Figure 16. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 7 (Swedish x Swedish).

Final plant height, rn ~~~~~~~~~~~~~~~~~~~~IIIIIIIII~IPII~~IIIIIIIIIIIII;III~IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII~IIIIIIIIIIIIII~IIIIJ - -I O-3 AC 1002 x AC 1009 -I m = - I- - -I - "- a. a. .. -" - :: -m - I -m - -m. - -" - = - 0.e - -m - - - ="m . = - -m - - n. -,mn. -..wa - \ -.Y# -" - -rn . . "8 0.12 m ". ". . -c- m - . rn -m ". &. &. ,mra rn TRBD~~PAIIUIIIIIIIIIIIB~IPPBBKFRPF~~P~~~~PPIIUE~ ~#;~I;~~~~~II~~~~B~~I~R~~~~III!LI~~IIIIIIIIIII~II~IIIIIIIBIIRRPII~IIIIBBI~ 6b $0 8 0 90 100 Percentage growth 35 days before the termination of the experiment Figure 17. The relationship between the total plant height and the percentage growth at five weeks before the termination of the experiment. Progeny 8 (SwedishxSwedish). Total height growth 1974, m

Percentage growth on July 11 Figure 18. The relationship between the total growth and the percentage growth on July 11, 1974 in a progeny from a cross between a north Swedish clone and a clone originating from the provenace Doha, Soviet union. The data refer to the fifth growing season.

This analysis seems to support the idea that growth on those two dates it seems that a the less laborious method I1 could be used measurement made at an intermediate date when investigations on a large scale have would have been more profitable for a study to be carried out. A prerequisite is, of of the present type. Detailed analysis of course, that the measurement during the the growth pattern under outdoor condi- growth period is carried out at an appro- tions have to be carried out to permit a priate time. general use of method I1 in selection work. The method I1 technique was applied to It would be advantageous if such a test some intra- and interprovenance hybrids could be performed at localities with dif- growing in a nursery at Bogesund (9 kilo- ferent light regimes. meters northeast of Stockholm). The plants were measured every 14 days during their active growth. The relationship between the 3.3 Total plant height-total branch length total growth and the percentage growth on June 27 and July 7 respectively, was studied. It might be speculated as to whether or not Also in this material a negative relationship the growth capacity could be attributed to predominated. However, the correlation a large extent to the size of the "photo- coefficients were in most cases non-signifi- synthetic apparatus". A very rough estimate cant. One example from this study is shown of the size of this characteristic would be in Figure 18. Based on the percentages of to measure the length of the branches. Therefore, such a measurement was carried Table 4. The correlation and regression out at the termination of experiment 11. coefficients for the relationship between The relationship between the values ob- total plant height and total branch length. tained and the total plant height was in- vestigated. The data obtained are compiled Type of progeny Correlation Regression coefficient coefficient in Table 4. As seen from this table the regression coefficients were positive with Swedish x Swedish the exception of those for progeny 0 1006x BD 2006. However, no very far-reaching Z 3008 x AC 1009 0.674** 0.192 AC 1002 x AC 1008 0.726** 0.264 conclusions could be drawn from such a BD 2006 x AC 2003 0.611* 0.143 limited material but it suggests that such measurements are worthwhile in future ex- French x Swedish periments. 0 1007 x AC 2003 0.293 0.043

French x French 0 1007 x 0 1008 0.943** 0.362 0 1008 x 0 1009 0.649 0.160 4 Acknowledgement

The present investigation was supported by with foreign countries, which is acknowl- a grant from the Knut and Alice Wallen- edged. We wish to express our thanks to berg foundation, which is gratefully ac- professor Enar Anderson, Stockholm for knowledged. One of us, Velichko Gagov, reading the manuscript. Finally we wish to visited Sweden as a stipendiate of the thank Mr Kjell Lannerholm for drawing Swedish Institute for Cultural Relations the diagrams in this paper. 5 Summary

The present investigation showed that the continued on material grown under outdoor variation in height growth within a full-sib conditions. family in many cases could be attributed to In the paper a method is presented that different photoperiodic responses of the permits a selection in large plant materials. individual plants. The range of variation In this method the total plant height (or was in many cases considerable (cf. Figure growth during a certain season) is plotted 12). Since the plants for the first eleven against the percentage growth at a certain weeks were exposed to continuous light and time before growth cessation. Thereby it then exposed to different night regimes, the is possible to distinguish plants which com- range of variation may have been in- bine early growth cessation with a good fluenced. Therefore, the studies have to be total height growth capacity. 6 Sammanfattning

Syftet med den foreliggande undersok- I mfinga av helsyskonfamiljerna visade sig ningen var att klarlagga fr5gan "Varfor tillvaxtperiodens langd ha stort inflytande vaxer vissa granplantor inom en helsyskon- p5 planthojden. Skogstradsforadlaren ar familj battre an andra?" Ett klarlaggande mest intresserad av att kunna valja s5dana har betydelse plantor som vuxit bra under en kort vegeta- tionsperiod. Ett satt att klara detta urval 1. for att vagleda fysiologerna betraffande ar att tillampa den teknik som betecknats mera sofistikerade undersokningar over som metod I1 i uppsatsen. I denna avsatts skillnader i tillvaxtkapacitet hos olika den totala planthojden mot den procentuella plantor. tillvaxten en viss tid fore tillvaxtavslut- ningen. I diagram (figurerna 12-17) kan 2. for att vagleda skogstradsforadlare be- man sedan latt valja de plantor som ager traffande urval for sticklingsforokning. de onskade egenskaperna. Literature

Burley, .I.1966. Genetic Variation in Seedling and suppressed red pine trees - Bot. Gaz. Development of Sitka Spruce, Picea sitchen- 124: 146-154. sis (Bong) Carr. - Forestry 39: 68-94. Krutzsch, P. 1975. Die Pflanzschulenergebnisse Dormling, I. 1973. Photoperiodic control of eines inventierenden Fichtenherkunftsver- growth and growth cessation in Norway suches (Picea abies Karst. und Picea obovata spruce seedlings - IUFRO Division 2 Ledeb.) - Royal College of Forestry, De- Working Party 2.01.4 Growth Processes. partment of Forest Genetics, Research Symposium on Dormancy in Trees Kornik, Notes 14: 1-64. September 5-9, 1973. Eanglet, 0. 1964. Proveniensvalets betydelse for Dormling, I., Ekberg, I., Eriksson, 6. and produktion och skogstradsforadling av gran Wettstein, D. von. 1974. The inheritance of - Svenska SkogsvBrdsforen. Tidskrift 3: the critical night length for budset in Picea 145-155. abies (L.) Karst. - Proceedings, Joint Lattke, H. 1973. Moglichkeiten der Steigerung IUFRO Meeting, S.02.04.1-3, Stockholm. der Holzproduktion bei Baumweiden und Pp. 439-448. Pappeln durch Anbau von Sorten mit langer Hagner, M. 1970. The intra-provenance cor- Vegetationsdauer - Beitrage fiir die Forst- relation between annual rhythm and growth wirtschaft 2/73 7. Jg. Pp. 84-88. of single trees of Pinus silvestris L. - Stud. Perry, T. O., Chi-Wu, W. and Schmitt, D. 1965. For. Suec. 82: 1-40. Height Growth for Loblolly Pine Pro- Kiellander, C. L. 1970. Studies on Populations venances in Relation to Photoperiod and in Picea abies (L.) Karst. with Special Re- Growing Season - Silvae Genet. 15: 61-64. gard to Growth and Frost Resistance - Schiitt, P. 1962. Ergebnisse einer Auslese vor- Thesis. 31 pp. wiichsiger Pinus silvestris - Samlinge aus Kleinschmit, 9. 1974. Consideration regarding dem Langtag - Silvae Genet. 11: 39-42. breeding programs with Norway spruce SylvCn, N. 1940. Ling- och kortdagstyper av (Picea abies Karst.). - Proceedings, Joint de svenska skogstraden (Longday and short- IUFRO Meeting, S. 02.04.1-3, Stockholm. day types of Swedish forest trees) - Svensk Pp. 41-58. Papperstidning 43: 317-324, 332-342, 350 Kozlowski, T. T. and Peterson, T. A. 1962. -354. Seasonal growth of dominant, intermediate,

Electronic version 0 Studia Forestalia Suecica 2002 Edited by J.G.K.Flower-Ellis