METSÄNTUTKIMUSLAITOKSEN JULKAISUJA COMMUNICATIONES INSTITUTI FORESTALIS FENNIAE MEDDELANDEN FRAN SKOGS FO R S KNI N G SIN S T I T UT E T I FINLAND M ITT E I LU NG E N DER FORSTLICHEN FORSCHUNGSANSTALT IN FINNLAND PUBLICATIONS OF THE FINNISH FOREST RESEARCH INSTITUTE PUBLICATIONS DE L'INSTITUT DE RECHERCHES FORESTIBRES DE LA FINLANDE 84 HELSINKI 1975 METSÄNTUTKIMUSLAITOKSEN JULKAISUJA COMMUNICATIONES INSTITUTI FORESTALIS FENNIAE MEDDBLANDEN FRAN SKOGS FO R S KNIN GSINSTI T UT E T I FINLAND MITTEILUNGEN DER FORSTLICHEN FORSCHUNGSANSTALT IN FINNLAND PUBLICATIONS OF THE FINNISH FORBST RESEARCH INSTITUTE PUBLICATIONS DE L'INSTITUT DE RECHERCHES FORESTIfiRES DE LA FINLANDE 84 TOIMITTANUT — EDITED BY OLAVI HUURI HELSINKI 1975 COMMUNICATIONS INSTITUTI FORESTALIS FENNIAE 84 Holopainen, Viljo. 1974. Risto Olavi Sarvas in memoriam.... 1— 10 Lukkala, Aino. 1974. Risto Sarvaksen kirjallinen tuotanto. List of publications ll 17 84. l Sarvas, R i s to. 1974. Investigations on the annual cycle of devel opment of forest trees. 11. Autumn dormancy and winter dor mancy. Tiivistelmä: Tutkimuksia metsäpuiden kehityksen vuo tuisesta syklistä. Syys- ja talvihorros I—lol 84.2 Leikola, Matti. 1974 Muokkauksen vaikutus metsämaan lämpö suhteisiin Pohjois-Suomessa. Summary : Effect of soil prepara tion on soil temperature conditions of forest regeneration areas in northern Finland 1— 64 84.8 Lähde, Erkki. 1974. The effect of grain size distribution of natu ral and artificial sapling stands of Scots pine. Selostus: Maan laji tekoostumuksen vaikutus männyn luontaisten ja viljelytaimis tojen kuntoon 1— 23 84.4 Parviainen, Jari. 1974. Havupuiden latvakasvaimen ja neulas ten vuotuisen kasvurytmin määrittäminen. Esimerkkisovellutus männyn jälkeläiskokeeseen. Summary : Determination of the an nual growth rhythm of the terminal leader and needles of coni fers. Application to a progeny test 1— 27 84.5 Mälkönen, Eino. 1974. Annual primary production and nutrient cycle in some Scots pine stands 1— 87 84.6 Rummukainen, Ukko. 1975. Männyn käpysadon ennustami sesta. Summary: On forecasting the cone crop of Scots pine .... 1 — 26 84.7 Raul o, Jyrki &Kos k i, Veikko. 1975. Erilaisten rauduskoi vu jälkeläistöj en pituuskasvu Etelä- ja Keski-Suomessa. Sum mary: Height growth of different progenies of Betula verrucosa Ehrh. in South and Middle Finland 1— 30 RISTO OLAVI SARVAS IN MEMORIAM 1908—1974 Risto Olavi Sarvas in memoriam Metsäntutkimuslaitoksen metsänhoidon professori Risto Olavi Sar vas, yksi Suomen metsätieteen näkyvimpiä henkilöitä, kuoli 66 vuoden ikäisenä huhtikuun 8 päivänä 1974. Risto Sarvas syntyi 10. 2. 1908 Porissa. Hän suoritti metsänhoitajan tutkinnon Helsingin yliopistossa 1931 ja väitteli metsätieteen tohtoriksi 1944. Sarvas toimi aluksi kolme vuotta valtion metsähallinnossa erilaisissa avustajan tehtävissä Pohjois-Suomessa, minkä jälkeen alkoi toiminta ylim män metsäopetuksen ja metsäntutkimuksen saralla: vuodet 1934-38 yli opiston metsänhoidon assistenttina tänä kautena kuitenkin sangen mer kittävä lukuvuoden kestänyt opintomatka Saksaan Ebersvvaldin metsä korkeakouluun ja 1941-45 rintamapalvelun ohessa yliopiston metsän hoitajana. Jo ennen toista maailmansotaa (1938-39) Sarvas ehti toimia lyhyen aikaa Metsäntutkimuslaitoksen ylimääräisenä tutkijana ja v. 1945 hän palasi laitokseen uudestaan tällöin assistentin tehtävään. V. 1953 hänet nimitettiin Olli Heikinheimon jälkeen metsänhoidon tutkimusosaston pro fessoriksi. Sitä ennen hän oli toiminut assistentin virkansa ohella kuusi vuotta myös Metsänjalostussäätiön toiminnanjohtajana. Risto Sarvaksen elämäntyö on omistettu sangen ehyesti tiedemiehen kutsumukselle. Hänen suurtyönsä metsänhoitotieteen piirissä on ennen muuta Suomen metsien pääpuulajien uudistamistapahtuman syvällinen ja monipuolinen kartoittaminen. Alkuna näille tutkimuksille olivat jo 1930- luvulla suoritetut Lapin kuloalojen kasvillisuutta ja taimiston kehitystä koskevat, tunnustusta saaneet ja lupauksia herättäneet opinnäytetyöt. Myöhemmin 1930-luvulla Sarvas ryhtyi osittain käytännön taholta Metsäntutkimuslaitokselle esitettyjen toivomusten johdosta - tutkimaan harsintojen vaikutusta Suomen metsiin. Näiden tutkimusten joista mer kittävin oli väitöskirja »Tukkipuun harsintojen vaikutus Etelä-Suomen yksi tyismetsiin» (1944) tulokset vaikuttivat osaltaan hakkuutapojen vähit täiseen tervehtymiseen 1940-luvun lopulta lähtien. Sotien jälkeen Sarvas palasi jälleen uudistumistutkimusten pariin. Seu rasivat mm. tutkimukset koivun uudistumisesta (1948) ja siemenpuuhak kuusta männikön uudistushakkuuna (1949). 1950-luvulla pääaiheeksi muo 4 dostuivat metsäpuiden kukkimista ja siemensatoa koskevat ongelmat. Tämän erittäin mittavan tutkimustyön tuloksena ilmestyivät monografiat koivun (1952), männyn (1962) ja kuusen (1968) kukkimisesta, siemenen muodostu misesta ja leviämisestä, julkaisut, jotka alansa syvällisinä perusteita luotaa vina esityksinä ovat saaneet tunnustusta yli koko maailman. Näiden tutkimusten päätyttyä Sarvas paneutui 1960-luvun lopulla myö häisen kautensa suureen tieteelliseen ponnistukseen: hän ryhtyi selvittämään puiden vuotuisen aikataulun, »periodin», etenemistä ja riippuvuutta ympä ristötekijöistä. Sarjan v. 1972 ilmestynyt aktiivia periodia koskeva tutkimus ja vähän ennen tekijän kuolemaa valmistunut talvihorrosta koskeva osa muodostavat laajan, huolella kerättyyn aineistoon nojautuvan tieteellisen oppirakennelman, joka jo tähän mennessä on saanut osakseen laajaa kansain välistäkin huomiota. Erityisesti kukkimista ja vuotuisperiodia koskevilla tutkimuksillaan Sarvas on saattanut tietämyksen Suomen metsien pääpuulajeista aivan uudelle pohjalle. Nämä tutkimukset ovat jo tähän mennessä antaneet lähtö kohtia monille käytännön sovellutuksille, virikkeitä uusille tutkimuksille ja kieltämättä myös tutkijoiden väliselle debatille. Kaikesta päättäen näiden suurtutkimusten hyödyntäminen on vielä aivan alkuvaiheessaan. On kuiten kin merkille pantavaa, etteivät vain tutkijat, vaan myös aikaansa seuraavat, valveutuneimmat käytännön miehet ovat näiden tutkimusten arvon oival taneet. Ne ovat suomalaisen metsätieteen arvokkaita pysyviä saavutuksia. Ne tarjoavat sitä paitsi puhuvan esimerkin siitä, kuinka omista, kansallisista tarpeista versonut tutkimus rikastuttaa myös kansainvälistä tieteen viljelyä. Nyt esitetyn tutkimussarjan rinnakkaistyönä syntyi (1964) myös laaja dendrologian yliopistollinen oppikirja »Havupuut», osaksi omiin tutkimuk siin ja matkahavaintoihin sekä huolelliseen kirjallisuuden seulontaan perus tunut puulajitiedon arvokas koostumiis. Risto Sarvaksen tutkijaprofiili on erinomaisen selväpiirteinen. Erityisesti metsien uudistamistapahtumaa ja vuotuista periodia koskevat tutkimukset valottavat sellaisinaan yhtä hänen tutkijakuvansa olennaista, kenties olen naisinta piirrettä: sinnikästä, määrätietoista pyrkimystä tiedon perusteisiin. Suhteellisen triviaaleista havainnoista tutkija eteni niissä yhä syvällisempään ja vaativampaan puiden elintoimintojen analyysiin. Jo tehtävään ryhtyes sään hänen täytyi olla tietoinen siitä, että edessä on pitkä, vaivalloinen tie, mutta sitä ei Sarvas säikähtänyt, vaan hän eteni hellittämättömästi, askel askeleelta huolella laaditun aikataulun mukaan loppuun saakka, niin että viimeinen julkaisu saatiin painoon muutama viikko ennen hänen poisme noaan. Tuntui jopa siltä, että häntä kiehtoivat vaikeat tehtävät, että hän niiden parissa tunsi aidommin toteuttavansa tutkijan kutsumustaan. Olisi kuitenkin virhe kuvitella, että nämä vaativat työt olisi toteutettu vain tahdon voimalla. Risto Sarvas kävi niihin käsiksi hyvin varustettuna. 5 Hän omasi alun alkaen suuret hengenlahjat sekä jo tutkijauransa alku vaiheessa myös hyvät biologiset perustiedot, joita hän kartutti jatkuvasti ahkeralla opiskelulla sekä kotimaassa että ulkomailla. Hän on esimerkki tutkijasta, joka menestyksellä soveltaa perustieteiden metodeja ja oival luksia metsätieteen ongelmiin. Kolmas piirre Sarvaksen tutkijankuvassa on tukeva empiirinen ote. Hän oli syvästi kiintynyt tutkimuskohteeseensa metsään, viihtyi siinä ja innostui siitä jopa niin, että unohti monesti koealoillaan ajan kulumisen, unohti että päivä oli jo muuttunut illaksi ja että hän itse, työtoverit ja tutkimus apulaiset tarvitsivat ruokaa ja lepoa. Risto Sarvaksen metsä oli antoisa luonnonihme, mielenkiintoinen ja kiehtova, mutta hän käytti taitavasti hyväkseen myös laboratoriota ja sen välineistöä. Hän liikkui ahkerasti kaikissa Metsäntutkimuslaitoksen kokeilualueissa, mutta hänen mieluisin työympäristönsä oli Punkaharjun koeasema, missä metsä rikkaana ja moni ilmeisenä ympäröi häntä ja missä varsin vaatimattomiin huonetiloihin oli onnistuttu saamaan välttämättömät tutkimusvälineet. Sarvas koki tutkimuksen arvokkaana tehtävänä ja valitsi tutkijan kutsu muksen joutuessaan valintatilanteeseen v. 1962, jolloin Metsäntutkimuslai tokseen perustettiin ylijohtajan virka. Hän oli kuitenkin toiminut ansiok kaasti laitoksen johtajana kuuden vuoden ajan (1956-62). Näin hän saattoi sekä tutkijan että tutkimusjohtajan kokemuspohjalta tarkkailla aikansa tiedepoliittista kehitystä ja keskustelua. Hän seurasi sitä kiinteästi ja osal listui siihen hyvin harkituilla puheenvuoroilla. Hän toimi tieteen hyväksi myös eräissä valtion komiteoissa ja toimikunnissa, erityisesti metsäntutki muskomiteassa (1957-60), valtion maatalous-metsätieteellisessä toimi kunnassa (1960-66) sekä Pohjoismaiden välisessä metsäntutkimuksen yh teistyöelimessä (1972-74). Hän oli myös monien tieteellisten seurojen aktii vinen jäsen. Tiedepolitiikassa Sarvaksen erityisen huomion kohteena oli tutkijayksilö, hänen toimintaedellytystensä ja työrauhansa turvaaminen sekä tutkijan työn mielekkyyden kokeminen aikana, jolloin tiedepolitiikka on voimak kaassa murroksessa, tutkija entistä kiinteämmässä hallinnollisessa ohjauk sessa ja jolloin siltä usein näyttää jonkinlainen näennäistiede saa julkisuudessa enemmän arvostusta kuin vakava syvältä luotaava tutkimus työ. Sarvas kuitenkin oivalsi tutkimustoiminnassa ajan haasteet. On tietyn lainen paradoksi, että hän, alkuaan tyypillinen yksilösuorittaja, oli Metsän tutkimuslaitoksessa milteipä ensimmäisten joukossa kokoamassa eri tutki musalojen edustajista työryhmää metsänviljelyn tutkijaryhmää - ja suunnittelemassa toimintamalleja ryhmätyölle. Risto Sarvas ei ollut rooli-ihminen. Tavanomaiset ajatusraiteet olivat hänelle yleensä vieraita eikä hän niitä seurannut, vaan muodosti käsityk 6 sensä itsenäisen, usein kauan kestäneen harkinnan tuloksena. Hänen kannan ottojaan erilaisissa päätöksentekotilanteissa sävytti monipuolinen lähtö kohtien erittely ja pohdiskelu. Näin hän on jättänyt arvokkaita vaikutteita ei vain tutkijana, vaan myös tiedepoliittisena ajattelijana. Metsäntutkimuslaitos, Suomen metsätiede ja metsätalous muistavat suu ren kunnioituksen ja kiitollisuuden tuntein Risto Sarvasta, poikkeuksellisen mittavaa tiedemiestä. Viljo Holopainen Risto Olavi Sarvas in memoriam Risto Olavi Sarvas, Professor of Silviculture at the Forest Re search Institute and one of the most outstanding personalities in Finnish forest science, died on April 8, 1974, at the age of 66. Risto Sarvas was born in Pori on February 10, 1908. He graduated in forestry from Helsinki University in 1931 and defended his Doctoral Thesis in Forest Science in 1944. During the first three years of his career, Sarvas held various posts in the State Forest Administration in Northern Finland, after which he began his teaching and research activities in the field of advanced forestry. Between 1934 and 38 he held the post of assistant lecturer in silviculture at the University, during which time he spent a valuable study year at the Eberswalde Forestry University in Germany. Between 1941 and 1945, as well as serving at the front, he was the University Forest Officer. Sarvas had already served for a short while (1938-1939) as a supple mentary research worker at the Forest Research Institute, and in 1945 he returned to an assistantship at the Institute. In 1953, he succeeded Olli Heikinheimo as Professor and head of the Department of Silviculture. For six years during the time of his assistantship he managed the Foundation for the Improvement of Forest Trees. Risto Sarvas's life was whole-heartedly devoted to the science of forestry. His principal contribution to silviculture was to further the understanding of the regeneration of the main Finnish tree species. The basis for this research was already laid in graduate work which he carried out in the 1930 s on the vegetation and the development of seedling stands in fire-devastated areas in Lapland. The great promise shown in this early study received immediate recognition. Later in the 1930 s partly owing to wishes expressed to the Forest Research Institute by practical foresters Sarvas started to investigate the effect of selected cuttings on the Finnish forests. The most notable of the results was his doctoral thesis on »The effect of the selected cuttings of heavy timber on the privately-owned forests in South Finland» (1944). His findings contributed towards a gradual improvement in cutting practices from the late 1940 s onwards. 8 After the war Sarvas returned to regeneration studies. Among other contributions, they led to a report on birch regeneration (1948) and on the seed-tree method of regeneration cutting for a pine stand (1949). In the 1950 s his interest was caught by a new set of problems, research into which gave rise to monographs on the flowering, seed formation and seed dissem ination of birch (1952), of pine (1962) and of spruce (1968). These thor ough and penetrating publications have found recognition all over the world. On completion of these studies Sarvas turned the main weight of his scientific effort to the investigation of the annual rhythm of trees and its dependence on environmental factors. The study of the active period of growth published in 1972 and the work on winter dormancy completed shortly before his death, form an extensive body of carefully assembled research material, the scientific value of which has already attracted widespread international attention. As a result of his studies into the flowering of trees and their annual rhythm, Sarvas has given an entirely new dimension to our knowledge of the principal tree species of Finnish forests. His investigations have provided the starting point for much applied work, have stimulated new research and have stirred controversy among investigators. The application of these large-scale studies is clearly still in an early stage. It is significant that not only scientific workers but also progressive forest practitioners have realized the importance of these studies. They are lasting achievements of Finnish forest science, which also show how research which has sprung out of national needs can enrich scientific culture at the international level. Side by side with his primary research, Sarvas published a substantial university textbook »The Conifers» (1964). This valuable compilation was based partly on the author's own research, partly on the keen observation made during his extensive travels and partly on a discriminating use of the literature in the field. Risto Sarvas presents an extremely clear-cut picture of a research worker. The image is most clearly defined by his investigations into forest regenera tion and the annual rhythm of tree development. They display his per severance, his tenacity and the resolute effort that he always made to penetrate to the fundamentals of knowledge. From relatively trivial observa tions he advanced towards an increasingly profound and demanding analysis of the life processes of trees. On embarking upon this mission, he must have known that the road ahead was long and exacting. Nevertheless he advanced along it according to a carefully planned schedule step by step until the end. His last publication went to press a few weeks before his death. It seemed as if difficult tasks held a fascination for him and that he could satisfactorily fulfil himself by accepting their challenge. 9 2 5668—7 5 Yet these exacting tasks were not carried out by will-power alone. Risto Sarvas was endowed with great mental ability and was well-equipped to tackle them. But he was aware of the constant need for industrious study both at home and abroad if his knowledge of biology was to be sustained and deepened. He is a good example of a research worker who successfully applies the methods and insight of the basic sciences to the problems of forestry. A further characteristic of the image of Sarvas as a research worker is his firm empirical grasp. He was deeply attached to the subject of his studies, the forest. He was happy in it and was carried away by it. On his sample plots he was known to forget the passing of time, to forget that day had turned into evening and that he himself, his co-workers and research as sistants needed food and rest. He was a frequent visitor to all experimental areas of the Forest Research Institute but the working environment that he loved most was the experimental station of Punkaharju. Here, where his extremely modest room was fitted with nothing but the indispensable re search equipment, he was surrounded by all the riches and varying ex pressions of the forest. Sarvas regarded research as a primary duty and when confronted with a choice, gave priority to it. Accordingly, in 1962, when a full-time director ship was established in the Forest Research Institute he chose research instead of administration. He had, however, commendably headed the institute for six years (1956-1962). This had given him the opportunity to review develop ments in scientific policy and to discuss them perceptively from his ex periences both as researcher and research leader. Sarvas served his science on a number of State committees and boards, especially the Forest Research Committee (1957-1960), the National Research Council for Agriculture and Forestry (1960-1966) and the cooperative body of Inter-Scandinavian Forest Research (1972-1974). He was also an active member of several scientific societies. Sarvas understood the challenge of his day and age in research work. In his approach to scientific policy, Sarvas devoted particular attention to the research worker as an individual and to ensuring his conditions of work in a peaceful environment, striving to make him realize the significance of research work during a period of great change in scientific policy-making, a period when the research worker is under increasingly rigid administrative supervision and when so it often seems some kind of pseudo-science receives more public appreciation than serious and fundamental investiga tion. It is something of a paradox that Sarvas, originally a typically in dividualistic worker, was almost among the first in the Forest Research Institute to form a team of representatives from various disciplines — the forest regeneration research group and to plan activity models for team work. 10 Risto Sarvas was not a man to follow the herd. Conventional lines of thought were foreign to him and he never followed them. He was given to forming his own opinions after extensive deliberation, comprehensive anal ysis and assessment of the premises. The results were always distinctive and personal. He has thus left behind a valuable impression as a philosopher as well as a practitioner of his science. Finnish forestry owes a great deal to Risto Sarvas. Members of the Finnish Forest Research Institute and his many friends and colleagues in all silvicultural walks of life will remember an exceptionally able scientist with feelings of great respect and gratitude. Viljo Holopainen Itisto Sarvaksen kirjallinen tuotanto List of publications Luetteloinut catalogued by Aino Lukkala Lyhennyksiä - abbreviations AFF = Acta Forestalia Fennica MTA = Metsätaloudellinen aikakauskirja Metsätaloudellinen aikakauslehti MTJ = Metsätieteellisen tutkimuslaitoksen julkaisuja Metsäntutkimuslaitoksen j ulkaisuj a Communicationes Instituti Forestalis Fenniae SF = Silva Fennica FF = Folia Forestalia 1937 Kuloalojen luontaisesta metsittymisestä. Pohjois-Suomen kuivilla kankailla suoritettu metsäbiologinen tutkielma. Referat: Über die natiirliche Bewaldung der Waldbrand - fläehen. Eine Waldbiologische Untersuchung auf den trockenen Heideböden Nord- Finnlands. AFF 46.1, 146 s. Havaintoja kasvillisuuden kehityksestä Pohjois-Suomen kuloaloilla. Referat: Beob achtungen iiber die Entwicklung der Vegetation auf den Waldbrandflächen Nord- Finnlands. SF 44, 64 s. 1038 Havaintoja Pohjois-Suomen kuloalojen kasvillisuuden kehityksestä. Luonnon Ystävä 42, s. 215—221. Ilmavalokuvauksen merkityksestä metsätaloudessamme. Referat: Über die Be deutung der Luftfotogrammetrie in unserer Waldwirtschaft. SF 48, 46 s. Perä-Pohjolan kuloaloista ja niiden luontaisesta metsittymisestä. Metsälehti 6: 31, s. 4, 5. 1939 Ilmavalokuvauksen käyttäminen metsätaloudessamme. Die Photogrammetrie in unserer Waldwirtschaft. Maanmittaus, s. 237—247. 1940 Yksi jaksoisen metsikön luontaisista kehityskausista. Die natiirlichen Entwicklungs perioden der einschichtigen Bestände. MTA 58: 8, s. 79—83. 12 1941 Heikkotuottoisia metsiä polttohiilipuun hakkuilla kunnostamaan. Metsälehti 9: 1, s. 5—6. 1942 Havaintoja Syvärin varren metsistä. Metsälehti 10: 4, s. 6. 1944 Tukkipuun harsintojen vaikutus Etelä-Suomen yksityismetsiin. Referat: Einwirkung der Sägestammplenter-ungen auf die Privatwälder Siidfinnlands. MTJ 33.1, 268 s. Tukkipuun harsintojen vaikutus yksityismetsiimme. Metsälehti 12: 18, s. 5, 6. Havaintoja kanervatyypin paljaaksihakkuualoilta. Observations from clear-cutting areas of Calluna type. MTA 61: 6/7, s. 122—126. 1945 Harsintahakkuiden vaikutus yksityismetsiemme tilaan. Metsälehti 13: 14/15, s. 10—11. Puuston tiheys metsikön tunnuksena. MTA 62: 3, s. 80—84. Metsikön siemensadon arvioimisesta. MTA 62: 12, s. 389 —393. Ilmakuvien käyttö metsätaloudessa. Metsäylioppilas, s. 3—5. 1946 Huomioita kaivospuun ja paperipuun määrämittahakkuilla käsitellyistä metsiköistä. Summary: Notes on stands cut for pitprops and pulpwood to minimum diameter. MTJ 33.2. 27 s. Harsintahakkuiden vaikutus metsiemme ikäsuhteisiin. Metsäylioppilas, s. 16—18. Koivun rodunjalostuksen mahdollisuuksista. Summary: Possibilities of birch breeding. Suomen paperi- ja puutavaralehti 28: 12, s. 188—191. 1947 Metsäpuiden rodunjalostuksesta. MTA 64: 1/2, s. 29 —32. Millaiset puut soveltuvat metsäpuiden rodunjalostuksen kantapuiksi. MTA 64: 10, s. 223—226. Hurudana träd lämpa sig sasom stamträd för skogsträdens rasförädling? Skogsbruket 17: 11, s. 304—308. Dendrologian kurssin työvihko. (Moniste). Dendrol. Editio 11, 31 s. Metsäpuiden rodunjalostuksesta. Työtehotietoa 10, 9 s. 1948 Metsän pintakasvillisuuden kuvaamisesta. MTA 65: 6, s. 186—190. Tutkimuksia koivun uudistumisesta Etelä-Suomessa. Summary: A research on the regeneration of birch in South Finland. MTJ 35.4, 91 s. Harsinnan ajatus kitkettävä ammattikunnastamme. MTA 65: 11, s. 325—328. Harvennushakkuista. MTA 65: 12, s. 363—367. Metsäpuiden rodunjalostuksesta. SF 64, s. 33 —41. Metsäpuiden rodunjalostuksesta. Suomen paperi-insinöörien yhdistyksen vuosikokouk sessa 17. 4. 1948 pidetty esitelmä. Suomen paperi- ja puutavaral. 30: 17, s. 287 —291 & Maatalous ja koetoiminta 3, s. 228—239. 13 1949 Puumaiset koivulajimme. MTA 66: 1, s. 9—13. Siemenpuuhakkuu männikön uudistushakkuuna Etelä-Suomessa. Summary: Seed-tree cutting as a regeneration method in Scots pine forests of southern Finland. MTJ 37.5, s. 1—43. Metsäpuiden rodunjalostus. Suuri Metsäkirja I, s. 180—189 ja Tapion Taskukirja, s. 137—142. Huonokuntoiset metsät. Suuri metsäkirja I, s. 358 —379. Ulkomaiset puulajit ja niiden viljelymahdollisuudet Suomessa. (Yhdessä Lauri Ilves salon f kanssa.) Suuri metsäkirja I, s. 156—179. Forest genetics in Finland. 111 World Forestry Congress, Helsinki. Summaries of the Papers. Section I, s. 44—45. Bulletin No. 4. 1950 Tutkimuksia Perä-Pohjolan harsimalla hakattujen yksityismetsien uudistumisesta. Summary: Investigations into the natural regeneration of selectively cut private forests in Northern Finland. MTJ 38.1, s. 1—95. The effect of light on the germination of forest tree seeds. Oikos 2: 1, s. 109—119. Gallringarna och grundbest&ndet. Skogsbruket 20: 3, s. 67—72. Forest genetics in Finland. Actes du Ille Congres Forestier Mondial. 3 rapports speciaux, s. 135—137. Jättiläishaapa Helsingin kaupungissa. Summary: A triploid aspen in the town of Helsinki. MTA 67: 3/4, s. 98—100. Kolme vuotta metsäpuiden rodunjalostustyötä Suomessa. Summary: Three years of tree breeding work in Finland. MTA 67: 6/7, s. 229 —232. Koivu suomalaisena metsäpuuna. Suomen luonto 9, s. 17—27. Metsäpuiden siemenen hankkiminen ja käsittely. Tapion taimitarhakurssien luennot, s. 33—45. (Rotaprint). Mitä harsiminen on ja minkä vuoksi se pilaa metsän? Metsätietoa 1, s. 3—6. (Liite Metsälehteen n:o 15). 1951 Tutkimuksia puolukkatyypin kuusikoista. Summary: Investigations into the spruce stands of Vaccinium type. MTJ 39.1, s. I—B2. Perinnöllisyystiede, metsäpuiden rodunjalostus ja metsänhoito. Metsämies 42: 12, s. 256—259. Raudusko vai hies vanerikoivuna parempi. Summary: Hetula verrucosa or B. pubescens - which is the better veneer-birch. MTA 68: 2/3, s. 51 —53. Ohjeita lehtikuusen käpyjen keräämiseen. Metsätietoa 1, s. B—l 6.8—16. (Liite Metsälehteen n:o 26). 1952 On the flowering of bireh and the quality of seed crop. Seloste: Koivun kukkimisesta ja siemensadon laadusta. MTJ 40.7, s. I—3B.1 —38. Pohjois-Suomen kuivien kangasmaiden ekologiasta. Summary: On the ecology of dry moss-lichen forests in North Finland. MTJ 41.1, s. 1—27. Maatilametsälöissä suoritettavat harvennushakkaukset. Maatalous ja koetoiminta 6, s. 155—165. 14 Puun tuoton lisääminen rodunjalostuksen avulla. Talouselämä 15: 50, s. 982. (Kirja-arvostelu). Jorma Lehtonen: Metsäkasvioppi. Luonnon tutkija 56: 5, s. 159 160. Enemmän huomiota metsäpuiden siemenen rodulliseen laatuun. Suomen työ 39, 2 s. 1953 Measurement of the crown closure of a stand. Seloste: Puuston latvusyhteyden mittaa minen. MTJ 41.6, s. I—l 3.1 —13. Ohjeita pluspuiden valitsemista ja ilmoittamista varten. Instructions for the selection and registration of plus trees. SF 80, s. 93—100, 112. Metsänhoitajien jatkokurssi VII, s. 93—100, 112. Fröodling. Skogsbruket 23: 5, s. 154—158. Siemenviljelys. Summary: Seed source plantation. MTA 70: 3/4, s. 73—76. Hyvä rotu hyvä sato. Metsäpuiden rodunjalostussäätiö, Helsinki, 4 s. 1955 Ein Beitrag zur Fernverbreitung des Bliitenstabes einiger Waldbäume. Zeitschrift fur Forstgenetik und Forstpflanzenziichtung 4: 4/5, s. 137—142. Investigations into the flowering and seed quality of forest trees. Seloste: Tutkimuksia metsäpuiden kukkimisesta ja siemensadon laadusta. MTJ 45.7, 37 s. Nykyhetken ajatuksia harvennushakkuista. Summary: Today's thoughts about thin ning. MTA 72: 10, s. 315—320. Jättiläiskoivu löytynyt myös meidän maastamme. Metsälehti 23: 21, s. 1, 6. Björkens förnyelsefr&ga. Skogsbruket 25: 1, s. 18—20. 1956 Investigations into the dispersal of birch pollen with a particular view to the isolation of seed source plantations. Seloste: Tutkimuksia koivun siitepölyn leviämisestä eri tyisesti siemenviljelysten eristämistä silmällä pitäen. MTJ 46.4, 19 s. Pari lisäystä kasvukairatekniikkaan. Metsämies 47: 1, s. 2—3. Ett par beaktansvärda detaljer vid användningen av tillväxtborren. Skogsbruket 26: 2, s. 43—45. Metsänhoidon tekniikka. Metsäkäsikirja I, s. 498—564. Puulajit. Metsäkäsikirja I, s. 454 —474. Metsäpuiden rodunjalostus. Metsäkäsikirja I, s. 475 —480 & Tapion taskukirja 13. p., s. 138—143. Visaseura. MTA 73: 12, s. 415—418. Metsäpuiden rodunjalostuksen oppikirja. Kirjallisuusesittely. MTA 73: 12, s. 416—418. 1957 Studies on the seed setting of Norway spruce. Meddelelser fra Det norske skogfor soksvesen 48, s. 529—556. Tutkimustuloksia kuusen siemensadosta. (Pääkohdat esitelmästä, joka pidettiin Suo men Metsätieteellisen Seuran kokouksessa 23. 1. 1957.) Metsälehti 25: 6, 7 ja 8. Skogsskötsel och växtförädling. Svenska skogsv&rdsföreningens tidskrift 55: 4, s. 16—22. Birch for plywood. Finland —England. Finnish trade review 100, 2 s, Jättiläisvisakoivu. Metsälehti 25: 18, s. 1, 6. 15 1958 Kaksi triploidista haapaa ja koivua. Summary: Two triploid aspens and two triploid birches. MTJ 49.7, 25 s. Tallens fröskörd och dess tillvaratagande. Skogsbruket 28: 1, s. 9—15. Visakoivun siemenen hankkiminen. Metsälehti 26: 4, s. 2 & Visaseuran tiedonantoja 1, s. 11—15. Jättiläisvisakoivu. Visaseuran tiedonantoja 1, s. 7—lo. Metsäntutkimuslaitos 40 vuotta. MTA 75: 7, s. 222—226. Tuloksia maamme visakoivikon inventoinnista. Metsälehti 26: 38. s. 4. 1959 Der nordische Urwald. Schweizerische Zeitschrift fur Forstwesen 110: 3, s. 124—135. Metsäpuiden rodunjalostus. Tapion taskukirja 14. p., s. 129—135. 1960 Metsänviljelyksessä käytetyn siemenen kotipaikan etäisyys viljelypaikasta. Summary: The distance of the provenance of seed used in forest cultivation from the place of cultivation. MTA 77: 6/7, s. 217—220 & Metsätietoa 1. Metsäntutkimuskomitean mietintö. (Puheenjoht. Yrjö Ilvessalo, jäseninä Erkki K. Kalela, Erkki Laitakari, Jarl Lindfors, N. A. Osara, Veikko Pohjanpelto, Risto Sarvas ym.) Komiteanmietintö 12. Summary: Report of the Forest Research Committee. SF 109. 83 s. 1961 Lapin suojametsien käsittelyohjeet. (Yhdessä Eino Oinosen ja Gustaf Sirenin kanssa) 23 s. (Moniste). Uuden enteilyä metsänhoidon tekniikan ja sen perusteiden alalta. MTA 78: 2, s. 55—56. 1962 Investigations on the flowering and seed crop of Pinus silvestris. Seloste: Tutkimuksia männyn kukkimisesta ja siemensadosta. MTJ 53.4, 198 s. (Lisäpainos v. 1974. Reprint in 1974). The development of the tree species composition of the forests of southern Finland during the past two thousand years. MTJ 55.11, 14 s. Männyn kukkiminen ja siemensato. Summary: The flowering and seed crop of Scotch pine (Pinus silvestris). MTA 79: 12 & Metsätietoa 3, s. 473—475, 479. Kustskogarnas särart och skötselproblem. Skogsbruket 32: 11, s. 185—193. Metsäpuiden rodunjalostus metsiemme kunnostusohjelmassa. (Metsäviikon yleis kokouksessa 4. 4. 1962 pidetty esitelmä.) MTA 79: 5/6, s. 199—203 & Metsälehti 30: 15, s. 9—lo. 1963 Problems of flowering and seed production. (FAO) World Consultation on Forest Genetics and Tree Improvement, Stockholm. August, 23—30, 1963. Section 8/2. 111, 9 s. (FAO/FORGEN 63—8/2). (Moniste). Metsäntutkimuslaitoksen työsaralta: Metsänviljelyn peruskysymyksiä selvitellään. MTA 80: 3, s. 96—99. 16 1964 Havupuut. Porvoo —Helsinki, 518 s. Kiitettävä laudatur. (Suomen Metsäyhdistyksen kesäretkeily). Metsänhoitaja 14: 7, s. 162—163. 169. 1965 Valtakunnallinen metsäpuiden siemenen tarve ja sen tyydyttäminen. Puumies 11: 3, s. 54—55, 57. Professori Erkki K. Kalela. In memoriam. (Suomeksi ja englanniksi. Text in Finnish and English). MTJ 59, s. 5—13. Kalela, Erkki K.. Kirjallinen toiminta. MTJ 59, s. 14—18. 1966 Metsäpuiden kehityksen vuotuinen periodi. Esit. 8. 10. 1965. Suomalaisen Tiedeakate mian esitelmät ja pöytäkirjat. Helsinki 1966, s. 239—259. Yisakoivikon perustaminen ja hoito. MTA 83: 8, s. 331—333. Metsäpuiden kehityksen vuotuinen periodi. Summary: Annual period of forest trees. MTA 83: 12, s. 501 —505 & Metsätietoa 3. Siemenhuollon taustaa. KMS Tapio, Tapion tiedotuksia kentälle 1, s. 15—16. Metsänhuollon taustaa. Metsänviljelyseminaari 14—19.3, 3 s. (Moniste). Pohjois-Suomen puurodut. Lapin kansa 35, s. 7. 1967 Viljelymetsä. Juhlaesitelmä Metsäntutkimuslaitoksen 50-vuotisjuhlassa Helsingin yli opiston juhlasalissa 24. 10. 1967. MTA 84: 10, s. 288—291 & Metsälehti 35: 43, s. I—2.1 —2. Climatological control of flowering in trees. XIV lUFRO-Kongress, Munchen 4. —9. September 1967, Referate 111, Section 22, Miinehen, s. 15—30. (Rotaprint.) Pollen dispersal within and between subpopulations; role of isolation and migration in microevolution of forest tree species. XIV lUFRO-Kongress, Miinchen 4. —9. September 1967, Referate 111, Section 22, Miinchen, s. 332 —345. (Rotaprint.) The annual period of development of forest trees. Sitzungsberichte der Finnischen Akademie der Wissenschaften, 1965. Helsinki 1967, s. 211—231. Metsäntutkimuslaitos 1917—1967. Summary: The Forest Research Institute 1917 1967. Metsäntutkimuslaitos 50 vuotta. Institutum Forestale Fenniae quinquagena rium A. D. 1967. MTJ 65.1, 67 s. Kulturskogen. Festföredrag vid Skogsforskningsinstitutets 50-ärsjubileum i Helsing fors Universitets solennitetssal den 24. 10. 1967, 8 s. (Moniste) & Skogsbruket 37: 11, 276—277, 298—301. 1968 Metsänparannus- ja hoitotöillä pohjoisessa vaikeuksia ja rajoituksia. (Suomen Metsän hoitoyhdistyksen keskustelutilaisuus 2. 4. Metsätalossa. Yhdessä Leo Heikuraisen kanssa). Metsälehti 36: 14, s. 4. Kaksikymmenvuotias metsänjalostuksemme uuteen vaiheeseen. (Haastattelu). Metsä lehti 36: 6, s. 3. 11. 1969 Investigations on the flowering and seed crop of Picea abien. Suomenkielinen selostus; Kuusen kukkimisesta ja siemensadosta. MTJ 67.5, 84 s. 17 1970 Genetical adaptation of forest trees to the heat factor of the climate. World Con sultation of Forest Tree Breeding, 2nd, Aug., 7 —16, 1969 in Washington. I, Rome 1970, s. 187—202. (FO-FTB-69-2/15). (Moniste). The annual developmental cycle of forest trees. lUFRO. Section 22. Working Group on Sexual Reproduction of Forest Trees. Proceedings of the Meeting at Varparanta, Finland 28. 5.—5. 6. 1970. 11, Helsinki 1970, 16 s. (Rotaprint). Temperature sum as a restricting factor in the development of forest in the Subarctic. Resume: La temperature globale en tant que facteur restrictif dans le developpement des forets subarctiques. UNESCO. Ecology and Conservation. I. Ecology of the Sub arctic Regions. Paris 1970, s. 79—82. Problems of tree improvement near the arctic and the alpine tree lines. Actas del Sexto Congreso Forestal Mundial. 11, Madrid 1970, s. 1587—1590. Establishment and registration of seed orchards. FF 89, 24 s. Metsänrajakysymys ja suojametsävyöhyke. Niin metsä vastaa. . . Metsät ja luonnon suojelu, s. 145—151. 1971 Tutkimuksella ei ole riittävää tietoa Lapin metsänviljelystä. Metsälehti 38: 10, s. 12. Siemenen itäminen alhaisissa lämpötiloissa. Metsäntutkimuslaitos, Rovaniemen tutki musaseman tiedonantoja 2. s. 68—74. (Moniste). 1972 Investigations on the annual cycle of development of forest trees. Active period. Tiivistelmä: Tutkimuksia metsäpuiden kehityksen vuotuisesta sykluksesta. Aktiivi periodi. MTJ 76.3, 110 s. Mitä on metsänviljelyn tutkimus. (Esitelmä Suonenjoen metsänviljelyn koeaseman vihkiäistilaisuudessa 18. 9. 1972). Metsä ja puu 89 (4) 11. s. 28—29. Metsäpuiden taimien talveentumistapahtuma. Metsäntutkimuslaitos, Rovaniemen tutkimusaseman tiedonantoja 3, s. 30—34. (Moniste). 1973 The annual developmental cycle of forest trees. lUFRO Working Party S 2.01.4. Symposium on Dormancy in Trees. Körnik, Poland. (Moniste). Olli Heikinheimo in memoriam 1882—1973. (Suomeksi ja englanniksi. Text in Finnish and English). MTJ 78, s. I—B. Dormansi I ja sen pituuden kokeellinen määrittäminen. Metsäntutkimuslaitos, Suonen joen metsänviljelyn koeaseman tiedonantoja 9, s. 27—33. (Moniste). Männyn kävyn koon vaikutus siemensatoon. Metsäntutkimuslaitos, Rovaniemen tutkimusaseman tiedonantoja 5, s. 48—49. (Moniste). 1974 Investigations on the annual cycle of development of forest trees. II Autumn dormancy and winter dormancy. Tiivistelmä: Tutkimuksia metsäpuiden kehityksen vuotuisesta syklistä. MTJ 84.1. 101 s. INVESTIGATIONS ON THE ANNUAL CYCLE OF DEVELOPMENT OF FOREST TREES II Autumn dormancy and winter dormancy RISTO SARVAS f 8. IV. 1974 TUTKIMUKSIA METSÄPUIDEN KEHITYKSEN VUOTUISESTA SYKLISTÄ Syys- ja talvihorros TIIVISTELMÄ HELSINKI 1974 ISBN 951 -40-0115 -X Helsinki 1975. Valtion painatuskeskus PREFACE This investigation into autumn dormancy and winter dormancy was con ducted abreast of the study of the active period forming part of the annual cycle of development of forest trees (Sarvas 1972). In the course of the inves tigation, I have become increasingly convinced that this cycle is an entity so well integrated that it seems hardly possible to investigate or survey any of its parts successfully if they are separated from the whole. It is entirely due to publication technique that the results of this investigation are issued in two separate volumes. Actually, a full description of this indivisible whole would require a third volume a survey of the geographical variation in each of the three main parts of the cycle. But the data collected so far do not provide an adequate basis for a general survey of this question and this part of the study may have to be postponed for years. Now that I am about to conclude my investigation of the annual develop mental cycle of forest trees, which has continued for more than ten years, I wish to express my thanks to all those of my fellow workers who have participated in the task. The investigation has involved a great amount of work in the field and laboratory, as well as numerous calculations. Mr. Olavi Helenius, Forest Officer, Mr. Veikko Silander and Mr. Jaakko Rokkonen, Forest Technicians, have supervised the field work. Mr. Alpo Luomajoki, Ph.L., has participated significantly in the investigation of meiosis. The laboratory work has for the most part been done by Mr. Pentti Manninen. Mr. Timo Ylitalo, Forest Technician, has been in charge of the data pro cessing and the other calculations involved. Miss Katri Kallio has drawn the diagrams and Mrs. Aino Lukkala has done the typing. The manuscript was read by Professor Max Hagman, Professor Paavo Juutinen, and Dr. Veikko Koski, to whom I am indebted for many valuable suggestions. The Finnish language manuscript was translated into English by Mrs. Marja Dethlefsen and the translation was checked by Mrs. Jean Margaret Perttunen, to both of whom I tender my thanks. Helsinki, March 15, 1974. Risto Sarvas Professor Risto Sarvas passed away on April 8, 1974 when this publica tion was still in the press. The proof has been read by his colleagues. CONTENTS Page Abstract 5 Introduction 8 Experiments in a controlled environment 11 Dormancy I 11 Definition of concepts and working hypothesis 11 Experimental determination of the rate of progress of dormancy I ... . 16 Experimental method 16 Experiments with seeds 17 Experiments with seedlings 25 Experiments with cut twigs 27 Synthesis of experimental results 28 Length of dormancy I 32 Dormancy II 33 Definition of concepts and working hypothesis 33 Experimental determination of the rate of progress of dormancy II ... . 33 Experiments with Larix sibirica 33 Experiments with Alnus incana 39 Synthesis of experimental results 42 Determination of the end of dormancy II 44 Length of dormancy II 47 Investigations in natural conditions 52 Introduction 52 Pattern of parcels in the course of development 56 Dormancy I 66 Dormancy II 73 Hardening and frost resistance 78 Review of literature 83 Concept of dormancy 83 Influence of environment on the onset and termination of dormancy .... 85 Discussion 89 Tiivistelmä 97 Literature cited 100 ABSTRACT The author has divided the annual cycle of development of forest trees into three main parts: The active period, dormancy I (autumn dormancy) and dormancy II (winter dormancy). The results concerning the active period have been published earlier (Sarvas 1972). The present investiga tion deals with dormancy I and dormancy 11. The chief aim here, as in the study of the active period, has been to clarify in what way the progress of the cycle depends on environmental factors. The progress of the cycle is essentially the outcome of a sequence of physiological phenomena in the apical and lateral meristems. The results obtained earlier (Sarvas 1967, 1972) suggested that during the greater part of the cycle the moisture factor in these meristems is more or less optimal. As the progress of the cycle was observed in trees growing under natural conditions, it was not deemed necessary or even desirable, at least at this stage, to try to include the change in day length (photoperiod or nyctoperiod) in the model to explain the functional principle of the cycle. The pattern of progress of the cycle seems to be controlled largely by tem perature and time. In investigations on the dependence of the cycle on environmental factors it was possible in all the three main parts to make use of concepts that were alike in principle and research methods that were similar. The equation which conveys, in concise form, the basic principle of the progress of the cycle at a constant temperature is as follows: in which C\ —> C 2 is the cycle interval between times t x and t 2 , T a constant temperature and v (T) the rate of progress at temperature T. Where the tem perature fluctuates, as in nature, the equation takes the following form: For the sake of practical calculations, for instance, for automatic data pro cessing programs, equation (2) can be approximated in the form (1) C, -> C 2 = v (T) (t 2 t,) (2) C x -> C 2 = J v (T(t)) dt h Risto Sarvas 6 84.1 in -which i is the ith temperature measurement when the time measured between C x and C 2 has been divided into n equal parts. The unit of a cycle interval has been defined as follows: the unit of interval is that interval through which the cycle progresses at a certain base temperature in one hour. This definition is applicable to all three main parts of the cycle, the active period, dormancy I and dormancy 11. The same tem perature could have been chosen as the base temperature for all three parts. However, for historical and practical reasons this was not done. The base temperature chosen for the active period was 2°C and the unit of this cycle interval was denoted p.u. (period unit). For dormancy I, 3.5° C was chosen as the base temperature and the unit of cycle interval was denoted ch.u. (chilling unit). The base temperature chosen for dormancy II was 2°C and the unit of cycle interval at this main part was denoted d.u. (dormancy unit). From the definition of the unit of cycle interval and formula (1), the rate of progress of the cycle at any constant temperature can now be reckoned from the equation where hP is the time required for the cycle to pass through a given cycle interval at the base temperature P, and hT the time required in the same cycle interval at temperature T. The result is expressed as c.u./h. (cycle units per hour). The values for hT at several different constant temperatures were deter mined experimentally for each of the three main parts of the cycle. These values were substituted in eq. (4) and the regression of the rate of progress of the cycle on temperature could be determined separately for each of the three main parts of the cycle. In general, these regressions seem to apply Avell to both the vegetative and the generative cycles. These three regressions embody the first main principle concerning the progress of the cycle. The experimental data led to the formulation of two additional main principles: 1. the progress of the cycle is an irreversible phenomenon and 2. each of the three main parts of the cycle has its i constant length expressed in its own units of cycle interval1); when the sum of the units amounts to this value, the cycle passes, without any special impulse, into the next main part. r) Accordingly, length, as here used, does not refer to calendar time but rather to time weighted with temperature. (3) C, 2 = Z v (T t) Atj i = l (4) v (T)=-^- C . U ./h h T 84.1 Investigations on the Annual Cycle of Development of Forest Trees 7 In this way the research concerned with the functional principle of the progress of the cycle crystallized into two main problems: 1. what is the form of the regression of the rate of progress of the cycle on temperature in each of the three main parts of the cycle and 2. how long is each main part (at the levels of the individual, the subpopulation and the population) ex pressed in the units of the particular main part of the cycle concerned. The results pertaining to dormancy I and dormancy II are seen in Figs. 5, p. 29, and 9, p. 43, and in Tables 13, pp. 71—72, and 10, pp. 49—50. Investigations carried out in nature were largely focused on anthesis (dehiscence of pollen sacs) in subpopulations (stands) of different tree species. At this particular phase of the active period, when sampling at subpopulation (stand) level, it was possible to use recording devices which afforded great methodical advantages. The results were not incompatible with the prin ciples presented above. In fact, they threw additional light on the details of the pattern of progress of the cycle. An observation of considerable importance is that in general, cells, in passing through a certain phase, do not show a normal frequency distribution but an irregular multimodal distribution. Thus they progress in »parcels» which form more or less separate groups at the start of dormancy II and continue to be observable until the end of the active period following dormancy 11. The irregularity resulting from the parcels renders almost all sampling in research on the cycle very exacting. The hardening phenomenon in forest trees, i.e. the development in autumn of a physiological stage resistant to frost, is actually beyond the scope of this study. Nevertheless, it has been possible to gain a rough idea of the general timetable of the hardening process and how it is related to the annual cycle of development. INTRODUCTION This study on the autumn and winter dormancies, like the earlier one concerning the active period (Sarvas 1972), is confined to the investiga tion of physiological phenomena clearly included in the annual develop mental cycle of forest trees. For instance, a state of dormancy that results from prolonged drought and is not annually recurrent, at least in the northern part of the temperate zone, is not covered. Attention is concentrated on dormancy in buds. Investigation of a dor mant stage possibly interrupting the activities of the cambium and the root had to be excluded from the study on account of the vast amount of work that would have been entailed. The so-called correlative inhibition (cf., e.g., Romberger 1963 and Leike 1965) occurring during the initial phase of development of buds is not considered to be a true state of dormancy and has likewise been disregarded. Similarly, dormancy in seeds, which has re ceived much attention in the literature on the subject, was not one of the main subjects of this study. However, because seeds are easier to use in experiments they have been given some attention. The annual cycle of development of forest trees is divided into three main parts: the active period, autumn dormancy, and winter dormancy. In a detailed study of the active period (Sarvas 1972), the beginning and end of this main part of the cycle were defined as follows: the active period begins when dormancy ends and ends when dormancy begins. This definition is admittedly loose, evidently because of lack of sufficient information. We have now come one step forward. Probably the most important finding made in the study concerning the active period is that the rate of progress of the active period depends almost exclusively on temperature and that the regression of the rate of progress of the active period on temperature is the same for all the tree species investigated and for all the different phases of the active period, provided that the variation of the other environmental factors falls within the range of variation of the natural habitat of the tree (population) concerned. Dormancy may be defined as those parts of the annual cycle of develop ment which generally coincide with autumn and winter in natural popula tions growing in natural habitats, and during which the temperature rela 84.1 Investigations on the Annual Cycle of Development of Forest Trees 9 2 8727—74 tions of the developmental processes are essentially different from those obtaining in the active period. Of course, this definition, too, is far from perfect. However, it provides an objective criterion for the determination of whether a given meristem is in the active period or in dormancy. Dormancy is clearly not a homogeneous state but composed of different stages. It has long been known that dormancy is more difficult to break in the beginning than towards the end of this stage. The resting condition is said to be deeper in the beginning than in the later phases. However, such a statement conveys little. It may be better to say that at the beginning a longer x) rest (the word rest in itself is, of course, not a very appropriate term) is required than later on. However, what is perhaps even more impor tant is that the main change in the pattern of rest is probably not a gradual process but rather a single step of fairly short duration. This last-mentioned view finds support in many observations at micro scope level. A particularly concrete example is provided by meiosis in the microspore mother cells of Larix species. The transitional phase between pachytene and diplotene, i.e. between the active period and dormancy 11, is short, particularly in the individual mother cell. Similarly, the transitional phase between diplotene and diakinesis, i.e. between dormancy II and the active period, is short. The same is true of the starch accumulating in the apical meristems of many tree species in the course of dormancy I; at the end of dormancy I the starch disappears during a short period. As in an earlier paper (Sarvas 1970), dormancy is here divided into two main stages, dormancy I (autumn dormancy) and dormancy II (winter dormancy). These terms were deliberately chosen to be as neutral as possible, because more descriptive ones would have tended to cause misunderstanding. In fact, a more neutral term would clearly be better than the word dormancy (for instance, the active period = stage CA (cycle A), and dormancy I = stage CB (cycle B), and dormancy II = stage CC (cycle C) but the time is probably not yet ripe for a terminological revision as radical as this. The model of dormancy taken as the starting point of the present study resembles the one presented by Pfeffer as long ago as 1904. Pfef fe r divided dormancy in buds into two stages, an autonomous and an aitionomous stage: »I Autonome Knospenruhe (auch häufig endonome, endogene, auto gene), Austreiben der Knospen wird durcli in den Knospen selbst liegende Ursachen gehemmt oder verhindert. II Aitionome Knospenruhe. Austreiben der Knospen wird nur durch äussere Ursachen (ungiinstige Klimabeding ungen) verhindert». Pfe ff e r's autonomous dormancy may be identified, x ) Without doubt, the number of research workers holding a similar view about this important principle is increasing. For instance, Witk o w s k a-Z u k (1969, p. 385) uses the expression »less deep or shorter-lasting dormancy». 10 Risto Sarvas 84.1 at least to some extent, with dormancy I as presented in this study, and his aitionomous dormancy with dormancy II of this study. However, in the course of this century it has become increasingly evident that all physio logical phenomena are the outcome of interaction between the genotype and the environment. In principle, there are no truly autonomous physiologi cal phenomena, i.e. physiological phenomena that are independent of the environment. The purpose of the present study is to map out preliminarily and roughly the progress of autumn dormancy and winter dormancy, particularly that of the physical and physiological (antecological) characteristics of these states. In the literature dealing with dormancy there often appear the terms dormancy release and chilling requirement. These terms are expressions of more or less precisely defined patterns of ideas relevant to dormancy. Evi dently, they could be applied to the active period, too. We could speak of release of the active period and of the heat requirement of the active period. This kind of play with ideas is beneficial, for it clarifies the concepts now under discussion. When dealing with the active period, I did not, however, use these terms but instead spoke of the progress of the active period ( = release of the active period) and of the length of the active period ( = heat requirement of the active period). Thus, the difference lies in terminology alone. One of the goals of the present study was to construct as simple a model as possible of the whole cycle (of the sequence of physiological events in the cycle) and, therefore, in discussing dormancy, concepts and terminology that resemble those used earlier for the active period have been used, as far as possible. Thus, the expression dormancy release is replaced by progress of dormancy and instead of chilling requirement I refer to the length of dormancy (cf. footnote on page 6). It is necessary to emphasize repeatedly the all too obvious fact that the annual cycle of development of forest trees, at least in temperate cli mates, reflects the adaptation of the trees and of the populations composed of them to the annual cycle of the local climate, i.e. the alternation of summer and winter. The trouble is that, because it is so obvious, this basic fact, despite its importance, has not always received due attention. One of the reasons for this may be that the term dormancy, in particular, has been used in reference to physiological phenomena that may not be connected, at least directly, with the annual cycle of development. EXPERIMENTS IN A CONTROLLED ENVIRONMENT Dormancy I Definition of concepts and working hypothesis Eor many tree species, especially at the onset of dormancy, exposure to cold is a prerequisite for the progress of dormancy; accordingly, somewhere between the end of the active period and the beginning of the next active period there is a stage in which development only progresses at low tempera tures. This part of the cycle is termed dormancy I. The classic experiment that establishes the existence of dormancy I is as follows. The experiment is started in late summer and continues till mid winter. At regular intervals during this period, twigs of certain individual trees are brought into a greenhouse maintained at a constant temperature between 10° and 18° C. Such treatment, known as forcing, induces flushing. As a rule, however, the twigs brought in before the autumn cold do not flush at all. Only after sufficient exposure to low temperatures outside can they be forced into flushing. Experiments of this type were made early in this century e.g. by Molisch (1909). Since autumn 1909 they have formed part of the present investigation. The results obtained in the autumns of 1970 and 1972 are seen in Tables 1 and 2. The investigations covered the vegetative as well as the flower buds. It is perhaps well to emphasize that, after a minimum, specific to each tree species, the length of the forcing period is of no significance. In principle, all the twigs brought in for the purpose of forcing could be examined at the same time, for instance a month after the last lot of twigs was brought in. However, if this procedure had been used, analysis of the twigs brought in first would have been badly hampered by the effects of the long forcing period (for instance, buds would have withered). Therefore, several examinations were made during the forcing period. Of importance is not the length of the forcing period but the maximum percentage of buds that can be forced into the active period. Different tree species were found to differ considerably in the cold required to terminate dormancy I, that is in the »chilling requirement» (Tables 1 and 2). 12 Risto Sarvas 84.1 Table 1. Development of twigs brought into the greenhouse (17° C) from the forest at Punkaharju on successive dates in autumn 1970. Each tree species involved is repre sented by two individual trees and from each of these trees two twigs were taken each time. Buds were examined on Oct. 19, Oct. 28, Nov. 6, Nov. 12, Nov. 20, Nov. 27, and Dec. 10, 1970. The length of dormancy I will be considered later on. However, at this point it is already well to mention that when the experiments described above were made, attention was attracted by the great topophytic variation dis played in the flushing of buds. As a rule, the buds on the older shoots of a twig flushed after a shorter period of exposure to cold than those on the younger shoots. Flushing was particularly late in the buds on the shoots of the current year. Earlier, Witkowska —2 u k (1969, p. 386) was among those who studied the wide topophytic variation during dormancy I. Let us now state the working hypothesis as follows: 1. At the end of its active period, a given meristem enters dormancy I; 2. In dormancy I. the M ntprinl Date and hour of day when the twigs were brought intc the greenhouse Sept. 28 Oct. 5 Oct. 12 Oct. 19 Oct. 26 Nov. 2 Nov. 9 Nov. 16 Xua tel lal 7.25 8.30 10.40 13.00 13.40 13.50 15.00 Maximum percentage of buds that reached in development the active period Alnus incana, local origin; I 'unkahar ju LXII Male flover buds ] o 1 o 1 50 ' loo ; 100 100 I 100 j 100 Betula verrucosa, local origin; Punkaharju LIV Vegetative buds 0 0 100 Male flower buds 0 0 100 Female flower buds 1 0 0 100 Larix decidua, origin unknown; Punkaharju 8C 1 Male flower buds | 50 | 0 0 1 Larix Gmelini, origin RSFSR, Sakhalin; Punkahar ju 9 Microspores | 30 | 0 1 loo j 100 ! 100 [ 1 Larix sibirica, origin unknown; Punkaharju 49 1 Microspores | 100 | 100 I 100 | 100 I 100 I 100 I 100 ; 100 Populus tremula, local origin; ! Kerimäki, Silvola (near Punkaharju) Microspores J 1 o 0 ! 0 ! 0 1 80 ; 100 Chilling unit sum (cf. pp. 28 —3 10) at canopy level, Betula pubeseens, Punkaharju XIV ; 180 I 312 I 413 ! 517 604 ] G37 I 638 : 660 84.1 Investigations on the Annual Cycle of Development of Forest Trees 13 Table 2. Development of twigs brought into the greenhouse (17° C) from the forest at Punkaharju on successive dates in autumn 1972. Each tree species involved is repre sented by two individual trees, and from each of these trees, two twigs were taken each time. Buds were examined on November 30, 1972 and January 12, 1973. cycle progresses only at relatively low temperatures; 3. As soon as the progress of dormancy I has terminated, the meristem enters dormancy II (where the cycle progresses at high temperatures, too, but the regression of the rate of progress on temperature is not the same as in the active period, cf. Sarvas 1970). AT ifpriil Date and hour of clay when the twigs were the greenhous brought into Sept. 26 Oct. 3 Oct. 10 Oct. 17 Ost, 24 Oct. 31 Nov. 16 ,'lctlLl Icll 10.00 J 9.50 7.50 7.50 8.00 8.00 1 15.45 Maximum percentage of buds that reached in development the active period Abies sibirica, origin unknown; Punkaharju 45 Microspores | | 0 | 0 | 54 1 100 1 100 Betula papyri]era var. neoalaskana, origin Canada, Alta., Cooking Lake; Punkaharju 118 Vegetative buds | | | 70 | 100 | 100 ! wo | 100 Belula pubescens, local origin; Punkaharju XIV Vegetative buds | 0 | 0 | 15 | 100 | 100 , 1 ! Betula verrucosa, local origin; Punkaharju LIV Vegetative buds | 0 | 10 | 50 | 100 | 100 1 loo | 100 Larix decidua, origin unknown; Punkaharju 80 Microspore mother cells .. | 50 | 100 | 100 | 100 | 100 [ 100 I 100 Larix Gmelini, origin RSFSR, Sakhalin; Punkaharju 9 Microspore mother cells .... | 0 I 100 I 100 I 100 I 0 | 100 I 100 Larix sibirica, origin unknown; Punkaharju 49 Vegetative buds | | 10 I 50 I 80 I 100 1 100 1 100 Pieea Abies, origin Finland, Lammi; Punkaharju LII Microspore mother cells .... | 0 | 0 | A | 50 | 100 | 100 I 100 Populus t. remula, local origin; Sääminki (near Punkaharju^ ) Male flower buds | 1 0 I 0 I 0 I 0 1 0 Chilling unit sum (cf. pp. 28—30) at canopy level, Betula pubescens, Punkaharju XIV [ 109 I 238 I 322 1 425 1 542 I 652 1 Risto Sarvas 14 84.1 This model immediately raises three questions: 1. What is the regression of the rate of progress of dormancy I on temperature? 2. How long is dor mancy I? and 3. Is dormancy I a reversible or an irreversible chain of phe nomena? These questions resemble those asked in connection with the active period, or, to be more precise, they have been framed in a similar way. Thus here, too, it seems possible to try to proceed by using similar concepts and experimental methods. The investigation, like that concerning the active period, has been re stricted to the reactions to environmental factors which are displayed by natural tree populations growing in their natural habitats. With this restric tion, it may be assumed, initially at any rate, that the progress of dormancy I depends exclusively on time and temperature. Let us further assume that at a constant temperature the progress of dormancy I is directly proportional to time. Then the rate of progress, v, is simply a function of temperature, T, i.e., v = v (T). The cycle interval through which dormancy I has progressed at a constant temperature, T, can thus be expressed by the following equation: in which d x and d 2 are two phases of dormancy lat times t x and t 2 , and dj —> d 2 the cycle interval between phases d x and d 2. The unit of cycle interval for dormancy I is defined as the cycle interval through which dormancy I progresses in 1 hour at a constant temperature of 3.5° C. This was chosen as the base temperature because, according to preliminary the progress of dormancy I is most rapid at this temperature, an even more important point being that between 3.0 and 4.o°C the rate of progress is least affected by temperature fluctuations. All the shortcomings attaching to the unit of cycle interval for the active period (Sarvas 1972, pp. 13 and 34) also apply to the unit of cycle interval for dormancy I, determined in this way. Below, this unit is termed the chilling unit and is denoted by the symbol ch.u. Where the temperature fluctuates, as it usually does in natural condi tions, the equation expressing the relation between the cycle interval and the temperature takes the form where T (t) is the temperature at time t, and v (T) the rate of progress of dormancy I at temperature T. For practical purposes, equation (2) can be approximated in the form (1) dj ->d 2 = v (T) (t 2 (2) d x d 2 = J v (T (t)) dt <1 (3) dj d 2 =Z v (TO At, i = 1 84.1 Investigations on the Annual Cycle of Development of Forest Trees 15 in which i is the ith temperature measurement counted from the beginning of the cycle interval d x -> d 2 when the interval has been divided into n equal parts. We now observe that for reckoning the length of a given cycle interval, the only unknown is v. It is obvious that v can be determined experimentally (cf. p. 16). However, the practical difficulties involved may be expected to be greater than those faced in the corresponding task concerning the active period. Above (as in the investigation of the active period), a set of concepts was used. The most important of these is the unit of cycle interval for a given genotype. However, it was not possible to define a unit common to all the genotypes, not even to those of the same species. It is obvious that in the examination of dormancy I (as well as in the examination of the active period), other concepts based on a different ap proach can also be used. In studying the different possibilities, the first thing to be considered is what can be measured. As a matter of fact, the possibilities in this respect are very limited, at least for the time being. What can be measured are: 1. The period of time required for a given genotype (or a given population) to progress through a certain cycle interval at different constant temperatures and 2. The periods of time required for different genotypes (or populations) to progress through a certain cycle interval at the same constant temperature. It is due to these limited possibilities that, whatever the concepts used, we ultimately end up working with relative values. Therefore the question arises of whether it is advisable to try to get round this fact, as has been done in this investigation, by defining the unit of cycle interval as absolute in form, when in actual fact it is not absolute. The examination of dormancy I and of the whole cycle could from the very start be based equally well or perhaps even better on the relative rates of progress of the cycle. However, as seen above, in this investigation another starting point was chosen, because it was hoped that the concepts that would arise from it would be easier to work with, even though the reasoning finally leads in a round about way to the same conclusion as would be reached more directly by using the relative rates of progress. The results of the experiments given in Tables 1 and 2 show that in several species the termination of dormancy I has required less exposure to cold in the microspore mother cells than it has in the vegetative buds of the same twigs. This implies that the length of dormancy I is shorter in the flower buds than in the vegetative buds. A study of the microspore mother cells involves measurement of the length of dormancy I at cell level, whereas the time taken for the flushing of the vegetative buds is measured at bud level. It is possible that buds do not flush at the point when 50 % of 16 Risto Sarvas 84.1 the primordial leaves have passed through dormancy I but rather when almost all the primordial leaves, including the apical ones, have proceeded from dormancy I to dormancy 11. Furthermore, flushing of a vegetative bud signifies not only that both dormancy I and dormancy II have come to an end, but also that the active period has already made considerable progress. Accordingly, it is highly uncertain whether accurate observations concerning the length of dormancy I in the vegetative buds can be made on the basis of flushing. The difference between the lengths of dormancy I in the micro spore mother cells and in the vegetative buds is not so great that the dis crepancy is significant. Experimental determination of the rate of progress of dormancy I Experimental method At this stage, almost the only thing we know about dormancy I is that, by definition, this part of the cycle does not, practically speaking, progress at temperatures above 10° C. This definition is based on observation, as was mentioned in the introduction, and must be used as a starting point when we seek to increase our knowledge. It is appropriate to begin the investigation of dormancy I by clarifying how the rate of progress depends on temperature. Experiments can be made on cut twigs (on the vegetative or flower buds on them) or on seeds or whole plants. Since the variation within the material is likely to be large, several replicates are necessary. The procedure is as follows. From a large quantity (for example, several hundreds) of twigs, seeds or seedlings several matched sets are formed and divided into the requisite number of treatments and replications of treatments. For example, it is fairly easy to arrange experi ments on 10 000—50 000 separate seeds. Such huge numbers greatly increase the accuracy of the results. As regards the physiological state of the experimental material, it is only necessary to know that the active period has terminated but that dormancy I has not been completed. This can be tested in a pilot experiment by forcing a representative sample of the material before the experiment is begun for long enough at a constant temperature of, for instance, 15° C. Generally, at this temperature, if the material or some part of it is not yet in dormancy I, it will soon attain this stage. On the other hand, should it have passed through dormancy I, it would progress to the active period and, as a result, the buds on the twigs or seedlings would flush or the seeds would germinate. 84.1 Investigations on the Annual Cycle of Development of Forest Trees 17 3 8727 —74 The important point is that in principle it makes no difference which phase of dormancy I the experimental material has attained, as long as it is not too close to the end of dormancy I. The practical difficulties encountered in carrying out experiments on dormancy I are largely due to lack of knowledge of any suitable, easily observable phases during this stage of dormancy. The same applies to dor mancy 11. Therefore, the effects of the different treatments (temperatures) given during dormancy I can be observed only after the experimental mate rial has proceeded to the active period following dormancy 11. The experiments are divided into two parts: treatment and forcing. The treatments comprise different constant temperatures and different treat ment periods. As a result of treatment, either all the material, or part of it or none at all progresses to dormancy 11. By forcing the material after the treatment at a temperature high enough to prevent further progress of dor mancy I and by prolonging the forcing for a sufficiently long time (long enough for the material to pass through dormancy II and reach the flushing phase in the active period), conclusions can be drawn regarding the propor tion of the treated material that has passed through dormancy I as a result of the treatment. Experiments with seeds Although dormancy I in the apical meristems and dormancy in seeds have some characteristics in common, they obviously also differ fundamen tally in many respects ] ). Often, their ecological functions are very different. Dormancy in seeds is not included in the annual cycle of development but forms part of the regenerative physiology of trees. In this study, experiments on dormancy in seeds have been made only for methodical reasons with a view to gaining a better understanding of the temperature relations of dor mancy I in the apical meristems. The relation between temperature and the progress of dormancy I can only be determined by measurement of the time required for some given experimental material to progress through a certain cycle interval belonging to dormancy I at different constant temperatures (provided that the other environmental factors are about the same at all the experimental tempera tures). This is where the difficulty lies, because, as pointed out above, we have, at present, no knowledge of any physiological phases in dormancy I whose time-course can be observed. However, if we take randomized matched sets from material (e.g. a seed lot) that has previously undergone uniform treatment, we can presume that when the experiments are begun all these *) E.g. dormancy in the apical meristems seems to have no counterpart in seeds. Hence, the term dormancy in seeds rather than dormancy I is used in reference to seeds. 18 Risto Sarvas 84. l sets are in the same physiological phase of dormancy I. And we shall take this phase (to be exact, the phase which 50 % of the experimental material has passed through when the experiment is begun) as the starting point for the cycle interval to be investigated. As the end point we shall take that phase in which 50 % of the material has passed from dormancy I to dor mancy II or to the active period. It is probable that investigations into the dormancy in seeds carried out in different parts of the world outnumber those made with any other plant organs. One reason for this is that dormancy in seeds is of considerable importance; for instance, in large seed-testing laboratories questions con cerning dormancy inevitably arise. The seeming ease with which large quantities of seed can be handled in germination tests and with which sta tistically well-planned tests with this material can be arranged have, without doubt, tempted research workers to use seeds as research objects. However, the seed forms a mysterious miniature world of its own which, despite the enormous amount of research concentrated on it, continues to be largely obscure. This involves a risk. In arranging experiments as well as in placing interpretations on them, the research worker is only too well aware that something essential may have been overlooked. Earlier investigations have revealed that dormancy in seeds is achieved in different species in different ways, which involve different parts of the seed, and that, depending on the parts involved, dormancy often varies in character. In the present study, investigation has been concentrated on the phenomenon known as endogenous seed dormancy (cf., for instance, Villiers 1972, p. 229). Seed lots in which there was reason to suspect that dormancy was entirely or partly due to one of the following causes have here proved to be unsuitable as experimental material: 1. Retarded embryo development 2. Seed coat relatively impermeable to water and gases 3. Some mechanical obstructions that limit embryo development. In investigating the general model of endogenous seed dormancy it seems appropriate to choose, as test objects, tree species having seed with rather a thin coat in which the embryo forms the major portion of the live seed, the endosperm's contribution being small. In Finland, the birches Betula verrucosa J) and B. pubescens are suitable for this purpose. For theoretical reasons, a third birch species, Betula nana, also growing autochthonously in Finland, deserves attention. The experiments were first begun with Betula verrucosa and Betula pubescens. It immediately became evident, however, that in the seeds of Betula verrucosa, dormancy was not well developed, x ) In the nomenclature of trees Re h d e r (1940) is mainly followed. In fact, Betula verrucosa Ehrli. is the only exception. 84. l Investigations on the Annual Cycle of Development of Forest Trees 19 whereas the seeds of Betula pubescens displayed a strongly developed dor mancy. Consequently, all experiments were made with seeds of Betula pubescens. The seeds of the birch are not seeds in the botanical sense of the word; they are nuts (or nutlets). Below, however, in accordance with common practice, the term seed is used, since there is no likelihood of confusion. When dormancy in seeds is investigated the first questions to arise are surprisingly difficult to answer, viz. when should the experimental seeds be collected, and how should they be extracted from the catkins (temperature) and stored (humidity). Failure to apply the right preparatory treatment may ruin the experiment before it starts. For instance, at a high temperature at the time of extraction dormancy might be lost, and a low storage tempera ture might gradually release it. The first experiments on dormancy in seeds were made in winter 1970/71. All the seeds used in these experiments, a total of several litres, were collected from a Betula pubescens tree growing at Punkaharju x), in the autumn, before the temperature started to drop below 10° C. Only catkins that were mature and had turned brown and partly opened were collected. This was a means to be sure that the active period had terminated and that the seeds were definitely in dormancy. Consequently, the seeds, when collected, were fairly dry and possibly less susceptible than moister seeds are to extraction at a high temperature. Extraction and final drying took place at room temperature (about 22° C). The seeds were stored at the same temperature. Treatment of the seeds was carried out in incubators at different tem peratures and for different periods. In each incubator, a thermoelement was connected to the multipoint recorder. Later on, the fluctuations in tempera ture that occurred during the treatment were determined from the tape in the recorder and the average treatment temperature was calculated. Gen erally, this deviated only a few tenths of a degree from the prescribed tem perature. With the particular needs of this investigation in mind, a germinator (Figs I—3)1 —3) was constructed, of dimensions suitable for ca. 200 birch seeds. The idea was that each treatment unit (replicate) should have its own small germinator. Each germinator consists of 2 parts, a water reservoir (a large-sized drinking glass) and a germination tray (made from a disposable plastic Petri dish). The Petri dish and its cover are glued back to back. In the centre of the tray so constructed is a hole 2 cm in diameter made by burning. The other details are the same as in the Jacobsen germinator: a pad of blotting paper (filter paper), an interpad of stockinet and, underneath this, a cotton pad from which a lamp-wick is suspended. x ) The latitude and longitude of the experimental areas and some data on the experimental stands are given in Table 11, pp. 53—54. 20 Risto Sarvas 84. l Fig. 1. The germinator used in experiments with seeds. The temperature under the bell jar was measured with thermoelements placed under the blotting paper or under a seed in such a way that the thermoelement was protected from radiation. 84.1 Investigations on the Annual Cycle of Development of Forest Trees 21 Fig. 2. Attachment of the thermoelement to the germinator. The thermoelement is led inside the germinator through a hole drilled in the side of the upper Petri dish functioning as a germination tray. The wire is glued to the floor of the tray and the end of the wire is bent upwards about 3 mm. Fig. 3. Experiment to study the dependence of the rate of progress of dormancy in seeds on temperature. Germinative phase at 17° C. Punkaharju, spring 1971. Photo by 0. Oskarsson 22 Risto Sarvas 84.1 The germinator facilitates the transfer of seed from the imbibition treat ment to the incubators and then to conditions for forcing, so that the posi tions of replicates could be randomized at each transfer. Before the seeds were subjected to treatment, care was taken that their moisture content was maximal. Hence, 24 hours before the germinators were placed in the incu bators the seeds were spread on the germination trays and immediately sprayed with water. During this imbibition treatment the temperature was 14° C. All the treatments were carried out in artificial light, the intensity of which ranged between 1 500 and 3 000 lux. In the incubators the relative humidity of the air varied between 75 and 85 %. The treatment temperature was measured with thermoelements connected to a multipoint recorder. The arrangement was made as stable as possible by gluing the wire from the thermoelement firmly to the bottom of the Petri dish; only a portion about 3 mm long at the tip of the wire was left unglued. This was then bent up at an angle of 90° with the tip touching the underside of a seed (Fig. 2). Comparative measurements indicated that the average temperature on the underside of the seed was O.9°C higher than the air temperature measured (with a thermoelement protected from radiation) in the incubator, outside the germinator. After treatment the seeds were transferred to a warmer temperature (14— 18° C). As mentioned before, the method used here is based on the assumption that dormancy I does not progress at temperatures above 10° C. An important point to be considered here is that at very high temperatures ( > 22° C) the dormancy of Betula pubescens seed is lost rapidly. On account of this, the germination temperature was kept approximately constant in each subseries of experiments, although it varied somewhat (from 14° to 18° C) between the different subseries. In addition to temperature, humidity was controlled by taking care that the relative humidity of the air in the germination room did not drop below 80 %. The seeds were sprayed with water every day. Inspections were made once a day, starting at 16. o o h. A seed was con sidered to have germinated when the radicle was at least 3 mm long. At each inspection, the germinated seeds were removed and counted. A note was also made of the exact time of day. The experiments were discontinued after 3 weeks. The average percentage of empty seeds in the Betula pubescens samples used in the experiments proved to be 32.4 % and the germination percentage of well-stratified unsorted seed (filled and empty seeds) 65 %. Subsequent compilation of the experimental results is seen in Fig. 4, which shows the results of one subseries of experiments 31/1972, conducted at O.5°C. The graph is drawn on frequency paper. The horizontal axis shows the treatment periods at O.5°C and the vertical axis the total germination 84.1 Investigations on the Annual Cycle of Development of Forest Trees 23 obtained by the end of a 3-week period at 17° C, expressed as a percentage of the highest value possible (65 %, cf. p. 22). In the diagram, each point represents the mean value of 3 replicates (comprising 200 seeds each, a total of 600 seeds). For each experimental procedure (position in the incubator, order of removal from the incubator, position in the warm phase) the rep licates were randomized anew. The experiment is largely based on the assumption that in the seed population under investigation the frequency distribution of the lengths of dormancy in seeds (the chilling unit sums required for termination of dormancy in seeds) is about normal. Fig. 4 shows that the scatter of plotted points can be well represented by a straight line. This may be taken to mean that the distribution really is about normal. As Fig. 4 clearly shows, the longer the treatment period, the larger the percentage of the viable seed that has germinated. For example, after a 6- day treatment, only 19. o % of the viable seed had passed from dormancy to the active period. A treatment period of 7.7 days was required for 50 % of the viable seed to pass from dormancy to the active period, and a treat ment period of 10.9 days was needed for 95 % of the viable seed to proceed from dormancy to the active period. The standard deviation of the frequency distribution is considerable; Fig. 4 shows this to be 1.9 days, which in this case means that the variation coefficient is 1.9/7.7 = 24.6 %. However, the most important thing is that Fig. 4 shows the time that elapsed in the observed cycle interval of dormancy at O.5°C. It is seen that 50 % of the viable seeds progressed from dormancy to the active period in 7.7 days after the beginning of the treatment. Since the beginning of the treatment was the starting point of the cycle interval under investigation, the average length of the entire cycle interval in question at this tempera ture is thus 7.7 days. Without doubt, the rate of progress of dormancy in seeds also depends on environmental factors other than temperature. The moisture content of the seed, for instance, is of particular significance (cf., for example, Hari and Lehtiniemi 1972). In carrying out the present experiments, an effort was made to eliminate this effect by keeping the humidity close to the optimal level. It is possible, however, that these attempts were not altogether successful. The considerable variation that occurred between the different replicates of a certain treatment may have been due to differences in humidity. The great significance of humidity should also be considered when the experimental results obtained here are applied to phenomena occurring in natural conditions. In the autumn and winter, when, from the point of view of temperature, dormancy in seeds has every chance to progress, the moisture content of the seed of several tree species is low. For instance, measurements 24 Risto Sarvas 84.1 Fig. 4. The regression of the cumulative percentage of seeds that have passed through dormancy on the treatment period (at O.5°C). The graph is drawn on frequency paper. The horizontal axis shows the treatment period and the vertical axis the percentage of the viable seeds that have ger minated, following treatment, after exposure to a constant temperature of 17° C for 3 weeks. The seed: Betula pubescens, tree No. L, Punkaharju, collected in autumn 1970. 84.1 Investigations on the Annual Cycle of Development of Forest Trees 25 4 8727—74 carried out at Punkaharju on October 13, 1971, indicate that the moisture content of seeds of Pinus sylvestris was only 9.5 %. Therefore, under natural conditions dormancy in seeds cannot be expected to progress as rapidly as it would if temperature were the only factor affecting progress. In the foregoing, the progress of dormancy in seeds has been examined only at relatively low temperatures ranging between —l° and 20° C. It is well known, however, that high temperatures also release dormancy in seeds. Pilot tests made in the course of the present study indicate that when Betula pubescens seeds are exposed to temperatures above 23° C, dormancy can be overcome within about 24 hours. The dependence of the progress of dormancy in seeds on temperature was investigated in five separate experiments conducted with Betula pu bescens seeds. The results are compiled in Table 3. The results of calculations of the rate of progress of dormancy are also compiled in this table; they will be discussed later. Experiments with seedlings Dormancy can be investigated in seedlings, as in seeds, under controlled environmental conditions, without separating the experimental material from a whole to which it organically belongs, as happens, for instance, when cut twigs are studied. Of course, very possibly, perhaps even probably, the general pattern of dormancy I in the buds of small seedlings differs in many respects from the pattern displayed by seeds or by buds of mature trees; but another possibility is that the pattern is essentially similar in all these living cellular tissues. Before we resort to more complicated models, this last-mentioned possibility has to be tested. The experiments with seedlings were arranged largely in the same way as those with seeds and the exposure of these two types of material to dif ferent constant temperatures was partly carried out simultaneously in the same incubators. After treatment the seedlings were grown in prickle boxes (natural light) in a greenhouse where the temperature fluctuated between 15° and 19° C. In the experiments, balled seedlings 2—3 dm in lenght were used. Those of Scots pine were 1M + 1A x ) roll seedlings and those of Norway spruce and birch were IM + 1A paper pot seedlings. Before the experiments, the seedlings were stored in a greenhouse where, at a height of 1.5 m, the tem perature fluctuated between 12° and 16° C. Unfortunately, during this 4- month period they were placed on the floor, where the temperature was lower. As a consequence, dormancy I in the buds had already progressed farther than was desirable at the start of the experiment. 1) IM + 1A = plants grown one season in a plastic greenhouse and one year outdoors. 26 Risto Sarvas 84. t Table 3. Results of experiments clarifying the rate of progress of dormancy in seeds. The initial point of the cycle interval investigated (d , -> d 2) is that phase of dormancy which 50 % of the seed had passed through at the start of the experiment, and the end point is that phase at which 50 % of the seed had passed from dormancy to the active period. When placed in the incubators for treatment the seedlings were almost bare-rooted, with their roots between damp sheets of newspaper. For forcing, the seedlings, planted in boxes containing fertilized (NKP fertilizer) milled peat, were placed on a table in the greenhouse. The seedlings were sprayed with water twice a day and inspected at the same time. As a rule, their development was slow. The first experiment, carried out with Betula pubescens seedlings, began on February 1, 1972. The buds started to flush at the beginning of March. However, accurate dating of the flushing of the buds proved difficult and, therefore, a later developmental phase was used, the date when the annual Xo. of f? 1*" Illumination tive Experi- ment Treat- ment d i —> d 2 cycle interval, days v(T) = 1» /h experi- ment ity | hrs - /o lux began on temper- ature °C h 3 . 5 h-p s " 3» 5/nT ? ch.u./h. Seed: Betula pubescens, Punkaharju, tree Xo. a, 1970 31 75 j 8 1 300 9. 3. 72 0.5 1.6 4.0 4.6 7.0 7.9 10.3 1 5.0 7.6 5.0 6.6 5.0 5.1 5.0 - 6.1 5.0 9.7 5.0 I 14.6 5.0 — 1.9 1.7 1.2 1.6 2.6 3.7 0.66 0.76 0.98 0.82 0.52 0.34 0.0 32 80 8 1 3 000 13. 4. 72 -1.0 3.1 6.9 4.8 i 14.4 4.8 5.2 4.8 1 7.5 5.0 1.5 2.2 0.33 0.93 0.64 33 85 | 10 3 000 8. 5. 72 6.6 7.3 4.8 10.4 4.8 11.7 4.1 4.5 0.46 0.41 Seed: Betula pubescens , Punkaharj u, tree No. a, 1970, remaining seed lot 34 85 j 10 3 000 29. 6. 72 3.6 4.8 7.8 8.3 9.0 4.8 4.8 4.8 5.4 4.8 11.3 4.8 14.5 4.8 46.5 3.0 3.4 7.5 9.7 32.0 1.0 0.89 0.43 0.33 0.10 Seed: Betula pubescens, Punkaharju, Sample stand No L, random sampling of 7 trees, 1971 35 85 ; 10 4 000 28. 7. 72 4.0 5.2 6.5 8.3 9.2 4.8 4.9 4.8 5.8 4.8 7.5 4.8 15.0 4.8 I 22.5 3.3 4.2 0.3 10.4 15.4 0.98 0.83 0.64 0.32 0.21 84.1 Investigations on the Annual Cycle of Development of Forest Trees 27 shoot reached a length of 1 cm. That is to say, if the shoot was less than 1 cm long, the observation was assigned a minus sign, and if its length was 1 cm or more, the observation was given a plus sign. The observations were then treated in the same way as in the experi ments with seeds. Frequency paper was used to determine the length of treatment time required at each temperature for the cumulative percentage of plus observations to rise to 50 % when the forcing time had been suffi ciently long (about 1 month). The results of this first pilot experiment are seen in Table 4. They will be examined in subsequent pages in connection with a consideration of all the results concerning the progress of dormancy I. Table 4. Results of experiments on the rate of progress of dormancy I in the buds of seedlings. The initial point of the cycle interval investigated (di -> d 2) is that phase of dormancy I which 50 % of the buds of the seedlings had passed through at the time when the experiment was begun, and the end point is that phase in which 50 % of the buds of the seedlings had passed from dormancy I to dormancy 11. Experiments with cut twigs An attempt was also made to study the dependence of the rate of progress of dormancy I on temperature by experiments with twigs cut from mature trees. In fact, it was with such experiments that determination of the regres sion of the rate of progress of dormancy I on temperature was started. However, the various difficulties that arose during these early experi ments thwarted these attempts. The treatment had no effect at all on the development of either the vegetative or the flower buds. Even now it is not clear what actually caused the failure of the experi ments. A contributory factor was,