THE PHYSIOLOGY AND INHERITANCE OF FLOWERING IN CARROTS By MICHAEL HUGH DICKSON AN ABSTRACT Submitted to the School for Advanced G raduate Studies of M ichigan State U niversity of A griculture and Applied Science in p a rtia l fulfillm ent of the req u irem en ts fo r the degree of DOCTOR OF PHILOSOPHY D epartm ent of H orticulture 1958 / A pproved Lr: .Cs&J.. ;■ ProQuest Number: 10008545 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest, ProQuest 10008545 Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code Microform Edition © ProQuest LLC. ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 MICHAEL HUGH DICKSON ABSTRACT A 10-week induction p eriod a t 40° F produced the m ost rap id and u n i­ form flow ering for the th re e v a rie tie s, Chantenay, h n p e ra to r and N antes. S to rter storage p erio d s of four and seven weeks re su lte d in slow er o r in ­ com plete bolting. Size of the sto red root was found to have little effect on flow er induction down to the m inimum of approxim ately 1 cm in d iam eter te ste d in these experim ents. Roots of le s s than 1. 2 cm average diam eter grew m o re slowly and produced seedstalks la te r than ro o ts which averaged 2. 3 cm o r m ore in diam eter. To produce anthesis as e a rly as possible, the optimum growing te m ­ p e ra tu re in the greenhouse following cold induction was 50 to 60° F followed by 70° F a fte r seedstalks becam e visible. T his p ro ced u re saved two to th re e w eeks in the tim e from planting in the greenhouse to an th esis. F o r selection of non-bolting plants a field exposure of approxim ately 650 hours to te m p e ra tu re s below 50° F w as found to be the m inim um to induce bolting in all plants w ith a genetic bolting tendency. A p erio d of approxim ately 140 days from seeding was re q u ire d to allow m ost potential b o lte rs to p r o ­ duce flow er stalk s. Reducing this p erio d by 15 days re su lte d in five to ten p e rc en t reduction in the num bers of b o lte rs in a heavily bolting line. S elec­ tion for non-bolting is ineffective if late b o lters a re not allowed tim e to develop in the fall. ABSTRACT - 2 MICHAEL HUGH DICKSON G ibberellin produced stem elongation and flow ering, esp ecially in lines w ith a genetic tendency tow ard the annual habit. It m ight be p o ssib le to use such chem ical trea tm e n t in selection for non-bolting esp ecially in w arm y e a rs. As a m eans of a c c elera tin g bolting in m atu re ro o ts in the greenhouse, g ib b er­ ellin did not cause sufficiently uniform flow ering to be reliab le . was no e a rlie r than w ith conventional cold induction. Flow ering Five tre a tm e n ts with 100 ppm of gibberellin a t 7 to 10 day in terv als sta rtin g a t the 6-8 leaf stage induced the m ost uniform re s u lts . F ew er than five tre a tm e n ts re su lte d in som e elongation and flowering, but the resp o n se was not uniform in all v a rie ­ tie s. M ore than five trea tm e n ts caused excessive stem elongation, often without the production of flow ers. While m uch of the genetic data indicate that inheritance of bolting is quantitative, th e re is som e evidence that annual flow ering can re s u lt from the action of genes acting in a dominant fashion. T his is evident in se v era l b a ck c ro sse s of non-bolting F^ p a re n ts to the bolting p a re n t in which the backc ro s s progeny w ere all b o lte rs. However, the environm ent has a stro n g effect on the ex p ressio n of the annual c h a ra c te r and to evaluate the genotype of different plants would re q u ire careful control of environm ent. It is a p p a r­ ent that by rig o ro u s selection the genes for annual growth can be elim inated and non-bolting stra in s obtained. THE PHYSIOLOGY AND INHERITANCE OF FLOWERING IN CARROTS By MICHAEL HUGH DICKSON A THESIS Subm itted to the School fo r Advanced G raduate Studies of M ichigan State U niversity of A griculture and Applied Science in p a rtia l fulfillm ent of the req u irem en ts for the degree of DOCTOR OF PHILOSOPHY D epartm ent of H orticulture 1958 ACKNOWLEDGEMENTS The author w ishes to e x p re ss his sin c ere appreciation and thanks to D r. C. E. P eterso n fo r his advice and supervision in planning the ex p erim ents; and in the p re p a ra tio n of the m anuscript. The w rite r w ishes to thank D r. S. H. W ittwer for his advice and suggestions on the experim ents and in the p re p a ra tio n of the th e sis. He also w ishes to thank D rs . G. B. W ilson and J. E. G rafius and L. W. M ericle for th e ir guidance in the p re p a ratio n of the m anuscript. TABLE OF CONTENTS Page INTRODUCTION.................................................................................. 1 REVIEW OF L IT E R A T U R E ............................................................ 2 Ecology of F low ering in Some Biennial V egetables . . 2 A cceleration of Flow ering in Biennials by G ibberellin . 10 Inheritance of the Annual V ersu s Biennial C h a rac te r. . 15 Pollination and F lo ra l Development of C a rro ts and C elery 17 EXPERIM ENTAL.................................................................................. Flow er Induction by Cold T reatm en t in C a rro ts. . . . 19 19 V arietal D ifferences in Cold Induction R equirem ents 19 Effect of Root Size on Seedstalk Induction . . . . 22 Effect of T em p eratu re on the Tim e Interval Between the F ir s t V isible Seedstalks and A nthesis. . . 25 F ie ld Induction as a M eans to E lim inate Plants with an Annual T e n d e n c y ...................................................................... 26 Selection for N on-bolting L ines in E arly August V ersu s E a rly S e p te m b e r.......................................... 26 E stim ation of Induction in the F ield on a D egree Hour B a s is ............................................... 28 G ibberellin in C a rro t B re e d in g ............................................... 32 Induction of Bolting in Chantenay and Im perator C a rro ts in the G reenhouse by G ibberellin . . . 32 TABLE OF CONTENTS CONT’D Page The E ffect of P a rtia l Cold Induction plus G ibberellin on F lo w e rin g ......................................................................... 34 The Use of G ibberellin to Induce Flow ering in the F ield 37 E ffect of G ibberellin on F ield Induction of Flow ering in C a rro t Plants D iffering in Bolting Tendency . . . . 40 G ibberellin as a M eans of Selection fo r Slow Bolting in ............................................................ the F ield 43 Inheritance of Annual V ersu s Biennial C h a rac te r in C a rro ts 45 DISCUSSION AND CONCLUSIONS............................................................ 52 Flow er Induction by Cold T reatm en t and V ernalization . . . 52 Induction of Flow ering in C a rro ts by G ib b e re llin .................... 56 The Inheritance of the Annual V ersus Biennial C h a ra c te r in C a r r o t s ............................................................ 60 SUMMARY.......................................................................................................... 63 LITERATURE C IT E D ..................................................................................... 66 1. INTRO D U CTIO N W ithin th e l a s t few y e a r s c a r r o t v a r ie ty im p ro v e m e n t h a s d e v elo p e d fro m m a s s s e le c tio n to s y s te m a tic b re e d in g . T h e in c r e a s in g i n te r e s t in c a r r o t b re e d in g is in d ic a te d by th e fo rm a tio n of th e N a tio n a l C a r r o t C o m ­ m itte e , an d a c o o p e ra tiv e p r o je c t b etw een s e v e r a l S ta te E x p e rim e n ta l S ta ­ tio n s and th e U n ited S ta te s D e p a rtm e n t of A g r ic u ltu r e . T he d is c o v e ry of m a le s te r i l i t y in c a r r o t s h a s m ad e h y b rid se e d p ro d u c tio n p o s s ib le an d h a s p la y e d a m a jo r io le in the in c r e a s e d in te r e s t in c a r r o t b re e d in g . A s a r e s u lt of in c r e a s e d in te r e s t in c a r r o t b re e d in g , th e n e ed h a s a r i s e n f o r in fo rm a tio n on th e b e s t m eth o d to p ro d u c e p ro m p t flo w e rin g in th e g r e e n ­ h o u se fo llo w in g fa ll h a r v e s t, in o r d e r to h a v e s e e d m a tu r e d and re a d y fo r p la n tin g in e a r ly s p rin g . In fo rm a tio n on th e in h e rita n c e o f a n n u al v e r s u s b i­ e n n ia l flo w e rin g in c a r r o t s w as n e ed e d a lo n g w ith m e th o d s to s e le c t tr u e b re e d in g b ie n n ia l lin e s . In v e s tig a tio n s c a r r i e d o u t h ad fo u r m a jo r o b je c tiv e s : (1) to d e te r m in e o p tim u m c o n d itio n s fo r p ro d u c in g a n th e s is and m a tu r e se e d a s soon a s p o s ­ s ib le follow ing fa ll h a r v e s t; (2) to fin d m e th o d s to e lim in a te p la n ts w ith g e n e s fo r an n u al c h a r a c te r ; (3) to in v e s tig a te th e p o s s ib le u s e s o f g ib b e re llin a s an a id to c a r r o t b re e d in g ; and (4) to stu d y th e in h e rita n c e of a n n u al v e r s u s b i ­ e n n ia l flo w e rin g . REVIEW OF LITERATURE Ecology of Flow ering in Some Biennial V egetables T h ere have been m any p a p e rs on the environm ental conditions re q u ire d to induce flow ering in biennials. G a ssn er (1918) was am ong the f ir s t to in v es­ tig ate the effects of cold trea tm e n t on flow ering and seed form ation. He ob­ se rv ed th at beets held in a cool greenhouse from January to A pril went to seed the following sum m er, while those held at 20° C did not. G assn er also found that cabbage, rutabaga and c a rro ts seem ed to depend fo r th e ir flow er form ation upon the influence of low te m p e ra tu re s. 1. Size of plant a t com m encem ent of cold induction. Thompson and Smith (1938) w orking with onion se ts stated that larg e se ts of 13/16 to 1/8 inch diam eter produced much higher p ercen tag es of seed ­ sta lk s than did m edium 5/8 to 3/4 inch d iam eter o r sm a lle r se ts. On c e le ry Paw ar and Thompson (1950) investigated the effect of age and size of p lant at tim e of exposure to low tem p era tu re . They observed that older c e le ry p lants a t the tim e of cold trea tm e n t went to seed fa s te r than young plants, but that the total tim e from seed to seed was sh o rte st when 2 month old plants w ere trea te d . However, plan ts tre a te d at any age would bolt eventually. The age of the plant was the im portant factor, and size had no effect on the ra te of flow ering. 3. A ccording to Boswell (1929), bolting in o v er-w in tered cabbage plants 6 mm o r le s s in d iam eter, in cre a se d only slightly with in cre ased size of plant. At a d iam eter of 6 to 7 mm th e re was a sharp in cre ase in the percentage of b o lte rs. Above 7 mm the in c re a se was s till m ore m arked. E a r lie r planting in the fall produced in c re ased bolting even if the o v er-w in tered plan ts w ere the sam e size as th e re appeared to be g re a te r physiological m atu rity in old er plants. It was concluded that te m p e ra tu re is effective in prom oting reproduction only a fte r p lan ts have accum ulated a sufficient weight of re s e rv e foods. 2. T em p eratu re and duration of flow er induction. In onion sets Thompson and Smith (1938) observed that se ts sto re d at 40 to 50° F produced the highest percentage of seed stalk s a s w ell as the few est m ark etab le bulbs. Sets sto re d at 32° F produced the b est bulbs, and while se ts sto re d at 60 to70c F also produced a few flow er stalks, the bulbs w ere much sh riv elled . F o r seed production a growing te m p e ra tu re of 50 to 60° F was found to be best. In creasin g the photoperiod by five hours over the norm al day length hastened seed stalk development, but under high tem p e ra tu re s of 70 to 80° F, seed stalk s did not develop under eith er norm al o r long days. Ito (1957) ob serv ed that onion bulbs larg e enough to form seed stalk s only form ed flow er buds when the m ean tem p era tu re was le s s than 5°C and the m ax i­ mum le s s than 10° C. Those at the sam e leaf stage, but in which growth was re ta rd e d by replanting, form ed flow er buds at a m ean of 10° C and a m axim um and m inim um of 13° C and 2°C. Length of low te m p e ra tu re p erio d needed fo r initiation was 4 to 5 days. S ta rrin g (1924), in Montana, found that a check in grow th of c e le ry due to cool w eather alone unfailingly caused seed production. He a lso found that a t longer p e rio d s of cold te m p e ra tu re the percentage of b o lte rs in creased . In co n trast, Thompson (1929) stated that a se rio u s check in growth following cold delayed subsequent seed stalk developm ent in c e le ry and m ight p rev en t it e n tirely . In young seedlings a fte r two weeks o r m ore a t 40 to 50° F p r e ­ m atu re seeding was likely. Plants grown in the greenhouse fo r two and onehalf m onths o r m ore at 50° F w ere likely to produce seed stalk s in the f ir s t season. A te m p e ra tu re of 70° F a fte r 30 days of cold prevented bolting except when the seed stalk was evident before exposure to th is tem p era tu re . M iller (1929) found that m ature cabbage plants bolted following two m onths a t 40° F, but produced only a few b o lte rs following the s h o rte r exposure of 15 and 30 days at the sam e induction tem p era tu re . With garden beets Chroboczek (1934) showed that the crow n o r growing point was the v ital a re a which had to be subjected to cold tre a tm e n t to induce bolting. He was able to produce bolting by winding a sm all ru b b er tube around the base of the petio les and circu la tin g cold w ater through it a t 43. 8°F. If the tubing was wound around the base of the root, no seed stalk production o c cu rred . He also noted that germ ination at low tem p era tu re did not induce flow ering in b eets la te r grown a t high te m p e ra tu re s. T h irty days at 40 to 50° F fo r young p lan ts induced flow ering when grown la te r at 60 to 70° F, but 70 to 80° F nul­ lified the cold tre a tm e n t. If the plants w ere grown for 60 to 90 days a t 40 to 50° F, the nullifying effect of high te m p e ra tu re s was not evident. C arolus (1933) sto re d beets and tu rn ip s in cold sto rag e at 32 to 40° F and in common sto rag e at 55 to 65° F. Cold storage te m p o ra rily delayed se ed stalk elongation com pared to common storage; however, cold storage p lan ts eventually produced longer and m ore vigorous flow er stalk s. The ra te of seed stalk elongation was a cc elera te d as the sto rag e p erio d in creased . Thus, both beets and tu rn ip s sto red th ree months reached 90 p e rc en t elon­ gation ahead of those sto re d one o r two m onths. Likewise, with p a rsn ip s, Sakr (1952) rep o rte d 100 p e rc e n t bolting r e ­ sulted when p lants four m onths old w ere subjected to 25 o r m ore days in sto rag e at 40 to 50° F, and subsequently grown at 60 to 70° F. With m ature F ren ch F orcing c a rro ts, Sakr and Thompson (1942) found that storage at 40° F re su lte d in a higher percentage of b o lte rs than storage at 35 or 50° F. Roots sto re d 15 days at 40° F and then grown in the greenhouse at 50 to 60*F produced 100 p e rc en t b o lters, while 30 days storage at 35° F and 60 days a t 50° F w ere re q u ire d to produce 100 p e rc en t bolting at a growing te m ­ p e ra tu re of 50 to 60° F. Storage for 60 days at 40° F was re q u ire d to produce 100 p e rcen t bolting when plants w ere grown in the greenhouse at 60 to 70° F. Dickson (1956) found that m ature c a rro ts of the v a rie ty Chantenay when sto re d fo r four o r eight w eeks at te m p e ra tu re s of 32 to 50° F and then grown in the greenhouse at 55 to 60° F produced 90 to 100 p e rc e n t seed stalk s. Roots grown a t 65 to 70° F, following eight weeks cold induction also produced 100 p e rc e n t seed stalk s, but th re e m onths la te r than those grown at 55 to 60° F. C a rro ts grown at 55 to 60° F, following four weeks cold induction bolted 2 to 3 m onths la te r than those sto re d 8 w eeks. He found no difference in bolting due to differences in storage tem p e ra tu re s between 32 and 50° F. In plants sto re d eight weeks and then grown at 70 to 80° F, only 34 p e r cent produced seed stalk s. 3. Optimum te m p e ra tu re s fo r flow ering following induction. With c a rro ts Sakr and Thompson (1942) concluded that a growing te m ­ p e ra tu re of 50 to 60° F was the m ost favorable for seed stalk developm ent. However, a fte r a p erio d of 80 days or m ore at 40° F, all plan ts would go to seed even at higher grow ing tem p e ra tu re s. F ish e r (1956) stated that sh o rt photoperiods of 12 hours before v e rn a l­ ization induced e a r lie r and m ore profuse flow ering in c a rro ts . p erio d s of 18 hours a fte r vernalization stim ulated flow ering. Long photo­ Low light inten­ sity (35 foot candles) during vernalization inhibited flowering, and v e rn a liz a ­ tion in the d ark was best. He also found that a tem p era tu re of 32° C fo r th re e days a fte r v ernalization reduced the percentage of flow ering p lan ts. F lo w e r­ ing was reduced even m ore if the 32° C trea tm e n t was p reced ed by 3. 5 days a t 21° C im m ediately a fte r vernalization. F o r c e le ry seed production, Thom pson (1929) recom m ended growing c e le ry at 55 to 60° F a fte r cold induction until seed stalk s sta rte d to grow, then com pleting growth at 60 to 70° F. While with cabbage, M iller (1929) found th at plants a fte r two m onths induction, flow ered a m onth e a r lie r at 60 to 70° F than when grown at 50 to 60° F. Peto (1943) found that tu rn ip s which showed incipient bolting w ere inhibited by te m p e ra tu re s above 65° F . 4. V ernalization and seedling induction. Chesnokov (1943) stated that in turnips, cabbage and c a rro ts , young seedlings chilled 50 days produced a higher percentage of seed stalk s than p lan ts grown from slightly germ inated seed subjected to the sam e cold t r e a t ­ m ent. Sakr (1952) failed to observe bolting in seedling p a rsn ip s grown at 60 to 70° F, following 15, 30 o r 60 days at 40 to 50° F . Sakr (1944) also re p o rte d that when turnip plants one m onth old w ere subjected to 30 days at 40 to 50° F the subsequent exposure to 50 to 60° F was m ore favorable for seed stalk p r o ­ duction than higher te m p e ra tu re s. A fter 60 days of cold storage at 40 to 50° F any growing tem p e ra tu re gave a high percentage of b o lte rs. He found that seed subjected to m o ist conditions fo r one month at 37° F produced 64 p e rc e n t b o lte rs if it germ inated, but only th re e p e rc en t if it had not germ inated. Plants put into the cold when ju st germ inated tended to bolt m ore than those put into the cold when one m onth old. A te m p era tu re of 53° F and continuous light on germ in atin g seedlings caused incipient bolting in 71 days. In m atu re plan ts 30 days at 40 to 50° F followed by growing at 50 to 60° F produced the m ost se e d e rs. Seventeen day old c a rro t seedlings (Sakr, 1942) subjected to one m onth at 40 to 50° F produced m ore b o lte rs at a growing tem p era tu re of 50 to 60° F than when grown at 60 to 70° F. At th ree growing te m p e ra tu re s the p ercen tag e of bolting was low. When seedling plants w ere grown for 135 days at 40 to 50° F 54 p e rc e n t bolted, while the sam e seedlings moved to 60 to 70° F produced 43 p e rc e n t b o lte rs. Supplem ental light in c re ased vegetative growth at low te m ­ p e ra tu re s , had le s s vegetative effect at high te m p e ra tu re s and no influence on seed stalk production. When c a rro ts (Dickson, 1956) ju st seeded up to 56 days old w ere subjected to 28 days at 40* F and then grown at 60 to 70° F with photo­ p erio d s of 8, 12 and 16 hours none was induced to bolt. Kumaki (1956) found that vern alizatio n of c a rro t seed by cold storage at 2 to 3°C fo r one, two o r th ree monhs had a c le a r effect on flow er bud d iffer­ entiation and bolting, although the v a rie tie s te ste d differed in th e ir resp o n se. In studies on vern alizatio n of rye, a w inter annual, G regory and P urvis (1948) and P urvis and G rego ry (1945, 1952) observed that a sh o rt p erio d of high tem p e ra tu re (30° C) de v ern alized grain, but that it could be re vernalized in a sh o rte r tim e than the initial vernalization. The degree of de v ernalization v a rie d in v ersely w ith the duration of the previous vernalization. T hree days at 35° C, a fte r four w eeks vern alizatio n at 1°C, re su lte d in 55 p e rc e n t d e v e r­ nalization. A fter six w eeks, only 16 p e rc e n t devernalization o ccu rred , and a fte r 12 weeks none o c cu rred . Also, the vernalized condition was stab ilized ag ain st re v e rs a l if a p erio d of growth at 15° C im m ediately p reced ed the high te m p e ra tu re . In general it ap p ears that m ost biennials re q u ire about 60 days of cold induction at about 40° F in o rd e r to obtain flow ering within a m inim um tim e from the s ta r t of the cold induction period. Many stra in s of the different sp ecies w ill bolt following a sh o rte r period, but these stra in s a re usually inclined to behave as annuals, o r to segregate som e plants having the annual habit. F o r the definitely non-bolting stra in s, 60 days ap p ears to be a m in ­ imum induction p erio d . Onions a re a com plete exception req u irin g a m uch longer storage, which is a r e s t p erio d ra th e r than a th erm al induction needed by other biennials. With th is one exception, an induction p erio d of le s s than 60 days in m any stra in s of biennials will re s u lt in uneven bolting, o c c u rrin g over an extended period, depending on the genetic natu re of the p lan ts. 10. A cceleration of F low ering in Biennials by G ibberellin Prom otion of flow er form ation by tre a tm e n t with definite chem ical compounds has so fa r been accom plished only under highly specific condi­ tions and in iso lated instances. Indoleacetic acid induces flow er form ation in the pineapple and litche, and according to W ittwer and Jackson (1950) m aleic hydrazide induced flow er form ation in young c e le ry p lan ts. ev er, in oth er plants the sam e compounds w ere inactive. How­ G ibberellin is the f ir s t substance found capable of inducing flow er form ation under non­ flow ering conditions in a larg e num ber of rep re se n ta tiv e s of the cold r e ­ quiring ro se tte form ing p lants. Lang (1956a, b) f ir s t suggested gibberellin m ight rep la ce the cold induction req u ire m e n ts in a biennial. He used the biennial stra in of Hyoscyamus n ig er to show that gibberellin was capable of inducing stem elongation and flow ering under conditions which norm ally p e rm it only the form ation of a leaf ro se tte on an ex trem ely shortened ax is. N orm ally the plant re q u ire s exposure to low tem p e ra tu re s, and subsequently long days, in o rd e r to flow er o r to elongate. Lang grew the plants at tem p era tu re s above 20° C, which is above the th resh o ld te m p e ra tu re of 17° C needed to induce flow ering. Lang applied the solution at a concentration of 10 to 50 m g p e r litre with an eye dropper to the c e n te r of the ro se tte . When a total of 30 and 80 ug (m icrogram s) p e r plant had been applied resp o n se s becam e apparent, while at 6 ug, no resp o n se appeared. Thus, the m inim al dosage for ro se tte p lants is between 6 ug and 30 ug, which is higher than the amount needed to stim ulate stem elongation in som e n o n -ro se tte p lan ts. L ang (1956a, b) explains his re s u lts on the b a sis of a qualitative dif­ feren ce between the two types of p lan ts. In n o n -ro se tte plants only cell elongation o ccurs, while cell division as w ell as cell elongation o ccu rs in ro se tte p lan ts. In the ro s e tte plants th e re a re p ra c tic a lly no intertiodes p re se n t, thus n ece ssitatin g cell division p r io r to stem elongation. Five w eeks a fte r trea tm e n t stem elongation had p ro g re s se d considerably, but it was not until th re e w eeks la te r that flow er buds w ere v isible. tro ls rem ain ed s tric tly vegetative. The con­ In som e c ase s with higher co n cen tra­ tions th e re w ere injurious effects and the tip s of a num ber of plants w ere killed. However, a ll those escaping injury flow ered. effect w as observed only under long days. The flow er inducing The stem s elongated under short days, but the plants rem ain ed vegetative. In fu rth e r studies Lang (1956c, 1957) found flow er induction possible w ith gibberellin in som e oth er cold req u irin g plants. in a ll species, and was not uniform in m ost. It could not be induced With the e a rly F re n ch F o rcin g and D anvers Half Long v a rie tie s of c a rro ts, 10 to 20 ug of gibberellin p e r day was applied from the tim e the plants had produced good sized ro o ts until an th esis o ccu rred . T his re q u ire d 9 to 25 weeks of application fo r F ren ch 12. F o rc in g and even longer fo r D anvers Half Long. A pplications of 10 ug p e r day produced the b e st re s u lts , and higher ra te s w ere le s s effective. In tu rn ip s, he did not obtain 100 p e rc en t bolting in six m onths, and p a rs le y responded even m ore slowly. InP etkus w inter rye, a fte r th re e to four weeks of trea tm e n t with solutions of 0.1 to 5.0 ug p e r litre , stem elonga­ tion w as enhanced, but flow er form ation was not affected. It was noticed also in tu rn ip s and c a rro ts that flow er form ation was rap id and uniform when the te m p era tu re was rela tiv e ly low, but not low enough to effect d ir ­ ect th erm al induction. W ittwer and Bukovac (1957), and Bukovac and W ittwer (1957) re p o rte d that gibberellin induced flow ering and m arkedly hastened seed production in Chantenay c a rro ts not subjected to cold induction. Plants seeded in May and grown continuously in a greenhouse flow ered in O ctober without being ex­ posed to te m p e ra tu re s below 60° F . In a second experim ent they used 10 o r 20 u g p e r plant weekly o r single doses of 50 to 100 ug applied to the growing tip, sta rtin g when the plants w ere larg e enough to be subject to cold induction. They found that flow er p rim o rd ia f ir s t appeared 90 to 100 days a fte r t r e a t ­ m ent of the m ature ro o ts. However, none of the tre a tm e n ts induced 100 p e r ­ cent bolting and flow ering was not uniform . They also found that co llard s with a stem diam eter of 0. 5 to 1 .0 cm w ere induced to flow er when tre a te d with gibberellin at weekly in te rv als until 100 ug had been applied. L ikew ise, stem elongation o ccu rred prom ptly in cabbage of the Golden A cre and F erry * s Round Dutch v a rie tie s, when the plants w ere grown at 10 to 13° C and 15 to 18° C, and f ir s t tre a te d when they had a stem d iam eter of 1 cm . A few of the gibberellin tre a te d plants flow ered at the higher te m p era tu re , while none of the controls elongated at a ll. At the low er te m p e ra tu re all the tre a te d plants flow ered and in the case of the Golden A cre v a rie ty two out of the four control plants also flow ered. ated flow ering. However, the trea tm e n t g reatly a c c e le r­ The flow ers appeared norm al and se t seed when pollinated. Kale, turnip and rutabaga responded with stem elongation at both te m ­ p e ra tu re s , but flow ered only at the low er tem p era tu re . Bukovac and W ittwer (1957) found that in m ature celery, stem s elongated following treatm en t, but only a few plants flow ered. L a te r they showed that a few p lan ts of the v a rie tie s Sum m er P ascal and Utah 10-B grown at 18° C flow ered a fte r 20 weekly applications of 100 u g p e r plant sta rtin g at the 12 to 15 leaf stage. It was noticed that flow ering only o c cu rred when the stem s had elongated to over 150 cm s. Blaney (1950) tre a te d Foxgloves with a total of 60 and 300 ug over a p erio d of 30 days. The p lan ts with the low er application produced seedstalks and flow­ e rs , while those tre a te d with the higher concentration produced sh o rt stalk s but no flo w ers. no flow ers. Likew ise, cabbage and C anterbury Bells produced seedstalks but In an experim ent by G askill (1957) conducted on sugar beets, two o r m ore tre a tm e n ts of gibberellin re su lte d in the production of seed stalk s in a ll tre a te d p lan ts, but the seed stalk s produced w ere not of the reproductive type. However, su g ar beet plants receiv in g four applications of gibberellin at 1, 000 ppm sta rte d a fte r a 43-day p erio d of phototherm al induction and continued long days, flow ered e a r lie r than when gibberellin was not employed. The plants so tre a te d produced m atu re seed within 25 weeks from the date of the o riginal seed planting. The phototherm al induction p erio d by itse lf induced a few seed stalk s but no flow ering. The non-bolting stra in of su g ar beets norm ally grown re q u ire s m ore than a 43day phototherm al induction to flow er, and w ell over six m onths to com plete its life cycle. In a review of the effect of gibberellin on econom ic plants, W ittwer and Bukovac (1958) concluded that am ong the many biennials which have specific cold plus long day req u ire m e n ts fo r flowering, th ere a re none in which both the te m p era tu re and the photoperiod req u ire m e n ts fo r flow ering can be rep laced by gibberellin. They su m m arize as follows: “ Biennials which have flow ered irre sp e c tiv e of day length have done so reluctantly, and have e ith e r a facultative o r no long day req u ire m e n ts. G ibberellic acid effectively a c c e le ra te s flow ering in cold req u irin g biennials only when; (a) the photoperiod r e ­ quirem ents fo r flow ering a re satisfied; (b) they a re m aintained at te m p e ra tu re s slightly above those induc­ tive to flow er form ation; (c) they a re tre a te d a fte r cold req u irem en t is p a rtia lly satisfied; o r (d) the biennial c h a ra c te r is weak. '* Inheritance of the Annual V ersu s the Biennial C h a rac te r In alm o st all sp ecies whose econom ic form we know as biennials, th e re a re to be found plan ts growing as annuals. T hese annuals in c a rro ts , beets, su g ar beets, cabbage, turnip, clover and other crops a re a problem and in ­ convenience, and a cause of considerable econom ic lo ss to the grow er. It h as been the aim of the plant b re e d e rs to develop v a rie tie s fre e from th ese b o lte rs. The re p o rt of Sutton (1924) on the inheritance of annual flow ering in cabbage was the e a rlie s t effort to estab lish the inheritance of annual habit. He found that the tendency for cabbage to bolt e arly is caused by a re c e s siv e c h a ra c te r, although not fixed o r uniform . to r s also play a p a rt. a ll This indicates that secondary fa c ­ In his c ro s s e s between annual and biennial stra in s, plants w ere biennial. In celery , E m sw eller (1934) found that p re m a tu re seeding in inbred lin es tended to be caused by a re c e s siv e c h a ra c te r influenced by secondary fa c to rs. The sugar beet has receiv ed m uch attention with re g a rd to annual flow ering, van H eel (1927), the f ir s t to re p o rt on inheritance, stated that h e re d ita ry fa c to rs fo r the annual w ere re c e ssiv e because: (1) In som e c ase s he o bserved a higher percentage of annuals following inbreeding. (2) G ro sses of s tra in s having a few annuals, and stra in s having many annuals produced few annuals in the F^. m any annuals. (3) G ro sses of two stra in s with m any annuals produced In co n trast, all la te r w ork on sugar beets showed that bolting was caused by a dominant fac to r. M unerati (1931) found that when stra in s of b eets cultivated fo r m any y e a rs as constant annuals o r biennials, w ere c ro s se d rec ip ro ca lly , the F^ plants w ere p ra c tic a lly all annuals. seg reg ated into annuals and biennials in a m onohybrid ra tio . The When com ­ p a re d with the p a re n ts, the hybrid annual was always la te r bolting than the annual p a re n t. M unerati’s re s u lts w ere confirm ed by Abegg (1936) and Abegg and Owen (1936), who observed a ra tio of 1:1 in the b ack cro ss of the annual hybrid (Bb) X the biennial p a re n t (bb) and the expected 3:1 in the F . 2 Annuals in the F^ and F^ progenies w ere slow er in seed stalk development than p lan ts from the annual p a re n tal stra in s . Abegg (1936) concluded that dom inance, although shifting strongly tow ard the annual side, m ay not be com plete, a s its ex p ressio n was dependent upon m odifiers and environm ental conditions. In sw eet clover, M elilotus alba, Smith (1927) found 1563 annuals and 479 biennials in F progenies of heterozygous p lants. He concluded that a Zi single dom inant gene determ in es the form ation of annual p lants in sw eet clover. The gene involved determ ines the tim e of m atu rity and did not in ­ fluence the dorm ant p erio d between vegetative growth and fructification. Additional evidence re p o rte d by C lark (1935) confirm ed that the annual habit was dominant over the biennial. In re c ip ro c a l c ro s se s of white flow ered annual sw eet clover with annual Hubam sw eet clover, he showed that the two stra in s p o ss e s s the sam e dominant m utant. A ccording to Stevenson and White (1937), when an annual m utant appeared in a dw arf branching type of sw eet clo v er it behaved as a single dominant to the biennial. It was the sam e m utation a s in Hubam. Johnson (1942) found that the annual habit of the v a rie ty Golden Annual, an annual of M elilotus suaveolens, was dominant over the biennial of the sam e sp ecies. Pollination and F lo ra l D evelopm ent of C a rro ts and C elery In o rd e r to study the inheritance of annual v e rsu s biennial flow ering in c a rro ts som e knowledge of the flo ral development is needed. In c e le ry E m sw eller (1928) observed that autogam y did not occur and that p ro tan d ry was com plete. He found that flow ers opening in the m orning lo st m ost of th e ir stam ens by evening, and the an th ers dehisced soon a fte r the flow ers opened. About five days elapsed from flow er opening until the styles w ere fully elongated, and two and one-half days before all the pollen had been shed. The stigm a was not receptive until five to eight days a fte r the flow er opened. Borthwick (1931) and Borthwick and E m sw eller (1933) found that the dehiscence of c a rro t an th ers and fall of stam ens o c cu rred before the stigm as becam e recep tiv e. A ccording to th e ir observations, fertiliz atio n o c cu rred a fte r the sty les se p ara ted from each other, and ju st before p etal fall. They attrib u te d failu re of seed set on caged single um bels to the fact that an th ers deh isced and the stam ens dropped off before the stigm as becam e recep tiv e. They found that hybrids could be obtained by w aiting until the a n th ers on an um bel dehisced and stigm as appeared recep tiv e before supplying pollen from another plant. They used the ro o t length, color and height of the ro o t above the ground to dem onstrate that all th e seed from a c ro s s of Yellow Belgium with D anvers Half Long was hybrid. During the co u rse of the c a rro t breeding w ork at M ichigan State U niversity it has been noticed that occasionally a single caged umbel will se t a few seeds, when flies a re supplied as a pollinating agent; but usually no seeds a re produced. EXPERIMENTAL F low er Induction by Cold T em p eratu re in C a rro ts V a rie ta l D ifferences in Cold Induction R equirem ents C a rro ts of the v a rie tie s Chantenay, Im p erato r and Nantes w ere seeded at the Muck E xperim ental F a rm on May 20, 1956, and the ro o ts w ere h a r ­ v ested on Septem ber 14. About 100 ro o ts of each v a rie ty w ere placed in sh a l­ low w ire-bottom ed tra y s in m oist sphagnum m oss and sto re d at about 40° F. Storage p erio d s of four, seven and ten weeks w ere used. At the end of each p e rio d 30 ro o ts of each v a rie ty w ere rem oved from storage and ro lle d in a m ixture of T etram eth y lth iu ra m d isu lp h id e^ and sand, and then planted in 6-inch pots, using a soil m ixture of loam, sand and muck. T hree re p lic a ­ tions of 10 plants each w ere planted and grown in the greenhouse until January 19 under 18-hour photoperiods with night te m p e ra tu re s of 55° F. On bright days the te m p era tu re was between 60 and 70° F . A considerable lo ss of plants due to A ste r Yellows re su lte d in a reduction in the to tal num ber of plants p e r treatm en t. Nantes tended to bolt much e a r lie r than eith er Chantenay o r Im perator (Table I and F ig u re 1). In Nantes th e re was 82 p e rc en t bolting following 4 weeks of cold storage, 96 p e rc en t a fte r 7 weeks, and 100 p e rc en t a fte r 10 w eeks. Chantenay produced 72 p e rc en t bolting following 4 weeks storage, and 100 p e r ­ cent following both 7 and 10 weeks sto rag e. In co n trast, Im p erato r produced —^Supplied under the trad e name, A rasan, by Dupont de Nem ours C o ., W ilmington, D elaw are. TABLE I P ercen t Bolting in Chantenay (Q , Im p erato r (I) and Nantes (N) C a rro ts A fter Storage of Roots at 40°F fo r P eriods of 4, 7 and 10 Weeks 2 Seven Weeks C I N 21 22 19 T otal Roots F our Weeks^ C I N 14 15 22 Date D ecem ber 7 14 0 27 5 10 45 14 14 0 41 21 14 73 19 14 7 50 36 38 2 21 13 59 84 10 43 20 59 19 57 40 30 65 72 January F e b ru a ry 2 4 Ten W eeks3 C I N 20 25 14 - - - - - 91 7 10 48 38 96 72 65 96 89 48 96 86 80 100 73 95 52 96 93 85 100 47 77 95 62 96 93 85 100 53 82 100 95 96 100 90 100 *Stored Septem ber 14, planted O ctober 12. 2 Stored Septem ber 14, planted November 2. o Stored Septem ber 14, planted November 22. 4 Includes p rim o rd ia evident on F eb ru ary 2. BOLTING 21. PERCENT NANTES Weeks of storage 4 7 10 y^ PERCENT BOLTING 80 Mean of me mree varieties / IMPERATOR / / / 60 X / / // /X / / yv y /T 20 1--1 7 DEC. / 4/*^ x / 40 CHANTENAY / // // » / / / / 1 / / 1 s • a / t / ^ / / • / 1 1 11 11 21 18 4 JAN. 1 / e—i---ia/—i— i----1— i— i 2 P FEB. 7 21 DEC. 4 JAN. 18 i i 2 P FEB. F ig u re 1. Bolting in th re e v a rie tie s of c a rro ts following 4, 7 and 10 w eeks of sto rag e at 40° F and grown at 55° F night tem p era tu re . 53 p e rc e n t bolting following 4 w eeks storage, 95 p e rc en t a fte r 7 weeks, and 90 p e rc e n t a fte r 10 w eeks. In a ll cases, as shown in F ig u re 1, the 10 w eeks sto rag e re su lte d in 90 to 100 p e rc en t bolting m ore quickly a fte r planting than w ith sh o rte r p e rio d s of sto rag e. tim e and lab o r in the greenhouse. Thus, the prolonged storage tre a tm e n t saved Following 10 w eeks of cold storage, a ll th re e v a rie tie s bolted at about the sam e tim e and m ore uniform ly than follow­ ing a s h o rte r storage period. E ffect of Root Size on Seedstalk Induction Chantenay c a rro ts w ere seeded on May 20, June 16, July 1 and July 14, 1956, at the Muck E xperim ental F a rm . On Septem ber 10, ro o ts from all plantings w ere harvested, and sto re d in m oist sphagnum m oss at 40° F. On Novem ber 22, they w ere rem oved from storage, ro lle d in A rasan and sand and planted in 6-inch pots. They w ere then grown in the greenhouse at 55° F night tem p era tu re . Table H and F ig u re 2 show the difference in e ase of induction fo r the four ro o t siz e s . The larg e ro o ts tended to bolt slightly e a rlie r. The in v erse sine tran so rm a tio n of p ercentages as suggested by Snedecor (1946) was used fo r a n aly sis. Roots from seeds sown July 14 produced few er b o lte rs than the e a r lie r plantings. significant. The difference in this case was highly Sm aller but still significant differences w ere observed in bolt­ ing production in plants from the la te r plantings. Plants from the July 1 TABLE H E ffect of Root Size of Chantenay C a rro ts on Seedstalk Developm ent Follow ing 73 Days of Storage a t 40° F. No. Bolting on Date Seeding Date and A verage Root Size* May 20 3.55 cm 18 roots June 16 2. 95 cm 21 ro o ts July 1 2. 30 cm 22 ro o ts July 14 1. 22 cm 27 ro o ts January 2 6 3 5 3 January 10 13 13 11 7 January 19 14 14 15 12 January 30 15 15 18 17 F e b ru a ry 7 16 16 18 18 F e b ru a ry 14 16 17 18 20 F e b ru a ry 25 16 18 20 20 M arch 10 16 18 21 21 T otal bolted 16 18 21 21 Percentage 89 86 96 75 All plants w ere h arv ested Septem ber 10, and sto red until Novem ber 22. 24. PERCENT BOLTING 100 A Seeded B Seeded C Seeded D Seeded (A 26 2 DECEM8ER 9 16 23 JANUARY 6 May 20 June 16 July I July 14 13 2 0 27 FEBRUARY Figure 2. Effect Of planting date of Chantenay carrots on seedstalk develop­ ment following 10 weeks storage at 40° F. planting bolted le s s than those from the May 20 planting. Roots from the June 16 planting also bolted le s s re a d ily than those from the May 20 p la n t­ ing. The higher incidence of bolting induced in la rg e r ro o ts w as even m ore pronounced in F e b ru a ry and M arch. The bolting p ercen tag es observed during the p erio d January 2 to M arch 10 and the ro o t siz es fo r the four planting dates a re su m m arized in Table II* It should be noted that the g re a te s t d iffe r­ ence in bolting was due to the m uch slow er induction of the sm a lle r ro o ts from the July 14 planting. E ffect of T em p era tu re on the Tim e Interval Between F ir s t V isible Seedstalks and A nthesis Plants from the experim ent on ro o t size w ere used to study the influence of growing te m p e ra tu re on flow ering. Plants from the sam e age groups and size groups which w ere selected and p a ire d as seedstalks f ir s t becam e apparent. One plant of each p a ir was grown at 55° F, and the other was grown at 65 to 70° F. The in terv al between f ir s t visible seed stalk and anthesis w as reco rd ed . In late F e b ru a ry and M arch , on cloudy days the 65 to 70° F house w as kept a t 60°F . T his probably delayed som e of the la te r plants in that house. In the 65 to 70* F greenhouse th ere was an, average of 32 days between the f ir s t visible evidence of seed stalk elongation and anthesis; while in the 55° F house the in terv al was 47 days, a difference of 15 days. ence betw een p a irs w as 19 days. The average d iffe r­ Thus, ra is in g the te m p era tu re following the app earan ce of seed stalk s saved two to th ree weeks in the tim e to re a c h a n th esis. Such a saving of tim e is v e ry im portant in producing m ature seed fo r planting in the field in A pril. The p lants in the cool house grew about 12 inches ta lle r than those in the w arm house before reach in g anthesis. F ield Induction as a M eans to E lim inate Plants with Annual Tendency Selection fo r N on-bolting L ines in E a rly August V ersu s E a rly Septem ber Seed from four hundred experim ental inbreds was planted May 1 in 12foot row s in the c a rro t breeding plots a s p a rt of the c a rro t breeding program a t the Muck E xperim ental F a rm . Counts on b o lters in these lines w ere m ade on August 7 and Septem ber 4, 1957. T his gave a com parison in the num bers of b o lte rs e a rly in August v e rsu s a month la te r in e a rly Septem ber. The four hundred lines included 152 which had one o r m ore b o lters by Septem ber 4. The b o lte rs observed on August 7 and Septem ber 4 a re rec o rd ed in Table 3H. In these 152 lines, 727 bo lters w ere observed on August 7, and 1479 on Septem ber 9. On August 7, 47 of the 152 lines had no b o lters, while by Septem ber 4, the sam e 47 lines had a total of 151 b o lte rs. T h irty -fiv e of the lin es had one to four b o lte rs, with a total of 56, while the o th er 12 p r o ­ duced 95 b o lte rs fo r a m ean of eight p e r line. Of the 56 lines with one to th ree b o lte rs on August 7, 13 had not produced any m ore b o lte rs by Septem ber 4. 27. TABLE HI C om parison of the N um ber of B olters in 152 D ifferent Inbred L ines of C a rro ts on August 7 V ersu s Septem ber 4 August 7 No. of L ines Total Bolters Bolters p e r Line 0 0 S eptem ber 4 Total Bolters 56 95 Bolters p e r Line 1-4 5-20 33 12 45 0 0 13 19 24 56 18 27 44 1-3 1-3 1-3 18 52 215 7 30 4-5 58 4-14 43 1 52 608 727 6-42 985 1479 6-55 1-3 no change 1-4 5-22 N ineteen w ere s till w ithin the range of one to four b o lte rs p e r line, but 24 lin es had five o r m ore p e r line. The rem ain in g 50 lines had four o r m ore b o lte rs in August, and by Septem ber 4 had an average in c re a se of 9.0 b o lte rs p e r line. It is apparent that harv estin g ro o ts in August is u n d e sir­ able from the standpoint of selecting against the bolting tendency. E stim atio n of Induction in the F ield on a D egree Hour Basis Chantenay c a rro ts of the two lines fAr with a tendency to bolt (20 p e r ­ cent in 1956), and tBTw ith no tendency to bolt, w ere seeded in 10-foot row s a t the E xperim ental Muck F arm on four dates, A pril 20 and 26, May 10 and 17, 1957. The two lines w ere seeded alternately, six row s of each line being p lanted on each date in two replications. The effect of the duration of low tem p era tu re s on p e rc en t bolting b e ­ cam e apparent by the end of July. In Table IV the re s u lts a re tabulated in p ercen tag es of bolting, along w ith the hours of a ir and soil te m p era tu re b e ­ low 50* F up to August 1. F ifty deg rees F was taken as an a rb itr a ry te m p e ra ­ tu re below which c a rro ts and other biennials a re generally expected to be induced to flow er. Bolting data from the plantings at the Muck E xperim ental F arm (Table IV) w ere tran sfo rm e d from percen tag es using the inverse sine tran sfo rm atio n as suggested by Snedecor (1946). In line A by Septem ber 10 th e re w ere signifi­ cantly m ore b o lte rs in the A pril 20 planting than in the May 10 o r 17 plantings. 29. r- r- m no t— h in p. 0 co C O C O <3 o m ^ no & CCDO CO • O ' t • CO • bo . d < (N C O i-H O CQ CD S3 r- m ON • • CO t"* CO co O i 0 bo <3 bJD a O • * • -H o O C O O MO O O o o o o « co in oo no * • • • CO • CQ CO bO d CljN CO ON OC OC 'H O * d 0 C M (N O -tCD-J CO I in N O On O CO CD O aj mh C O © CO O in £o o no co m on m xr -o1 co CO On Is- On oCO n mO No Oo ^n C i-H NO CO co NO ""t1C O CO ON CO ON ON NO oo CO in N c-O NO oco CO ON on to C O Ui0 r3 m On CO n O 2 fc u •H <3 bo d 0 £ ° O NO co, co , , oi—r r'—r ^ ^ XJ 0 C O d <3 o TJ CO 3+0-> d aj ON O N COf) IN^ H d 5H d aj 3 3 1<1 S' <3 <3 0 >N ^ >N nd ■Mi01 5 CO X 0I nJ oo 32 m CO ^JN 0 p? d CQ X0J 0 CO dS CO CO (3) Figures not spanned by a continuous line differ at the one percent level. Duration of Cold Exposure and Resultant Percent Bolting in Carrots at the Muck Experimental Farm • 30. L ikew ise, those planted A pril 26 bolted significantly m ore than those p lanted May 17. However, in the slow bolting fBf line th e re w ere no significant differences in bolting. T here w ere wide v ariatio n s between the two rep licatio n s, re su ltin g in a larg e e r r o r te rm , thus n ecessitatin g la rg e d ifferences fo r s ta tistic a l significance between planting dates. In 1958 a s im ila r experim ent was p e rfo rm ed to obtain fu rth er in ­ form ation reg ard in g the effect of the spring induction p erio d on bolting. It w as hoped that the num ber of hours below 50* F n e c e ssa ry fo r induction and subsequent selection m ight be established. A s tra in closely re la te d to the TA' line of 1957 was used, plus com ­ m e rc ia l Nantes and Chantenay stra in s, which produced over 15 p e rcen t b o lte rs in the v a rie ty t r ia ls in 1957. The slow bolting line TBf used in the 1957 experim ents was again used in 1958. Seed was planted at the Muck E xperim ental F arm on A pril 9, 19 and 30, and May 12. Six 12-foot row s of each v a rie ty w ere seeded on each date in th ree replications of two row s p e r v ariety . Bolting percen tag es rec o rd ed w ere tran sfo rm ed using the in ­ v e rs e sine fo r p urposes of s ta tistic a l analysis. Table IV su m m arizes the duration of cold exposure and bolting in c a r ­ ro ts at the Muck E xperim ental F arm in both 1957 and 1958. The num ber of hours below 50° F at the soil surface and at four feet above the su rface is also reco rd ed . The fig u res given for p ercen t bolting a re from populations of 200 to 500 p lan ts. On August 2 highly significant v a rie ta l differences in bolting had developed, and th e re w ere significant differences between planting dates fo r all four v a rie ­ tie s . By Septem ber 13 a v ery c le a r p a tte rn had developed. T here w ere no significant d ifferences in p e rc e n t b o lters between the f ir s t and second p la n t­ ing in any v ariety, but in ev ery v a rie ty th ere w ere significant differences betw een the second and th ird seedings and in Nantes between the th ird and fourth seedings. In all case s the two e a rly plantings produced m ore b o lte rs than la te r plantings. The second planting receiv ed 639 hours below 50° F at a ir tem p era tu re and 669 hours below 50° F a t the soil surface. The la tte r is probably the m ore re lia b le figure a s to te m p e ra tu re effect on the plant. The f ir s t planting r e ­ ceived 802 hours below 50° F at the soil surface, as is shown in Table IV. T his e x tra exposure to cold caused no in cre ase in bolting, while the th ird planting w ith 497 hours below 50* F had significantly le ss bolting. Some plan ts esp ecially in the f ir s t planting w ere lo st due to fro st, and they m ay have left su rv iv o rs which germ inated late and as a re s u lt w ere effected by about the sam e amount of cold as the second seeding. G ibberellin in C a rro t Breeding Induction of Bolting of Chantenay and Im perator C a rro ts in the G reenhouse by G ibberellin T his experim ent w as designed to determ ine the effect of age and size of plant on flow er induction as influenced by the u se of gibberellin. Seed of Chantenay and Im p erato r v a rie tie s w ere sown on O ctober 27, 1956 in fo u r-inch pots, and p lants w ere tran sp lan ted to six -in ch pots January 10. The stra in s used had produced v e ry few b o lters under field conditions in 1956. F ifty pots of each v arie ty w ere grown at a te m p era tu re of 55 to 65° F until M arch, a fte r which the day tim e tem p e ra tu re s w ere often over 65° F. Plants w ere sprayed a t tw o-w eek in terv als with an aqueous solution of g ib b erellin at 100 ppm w ith 0.1 p e rc en t Tween 20 added. Eight plants of each v a rie ty w ere tre a te d at in te rv als of two weeks, sta rtin g on January 3. F o u r plants receiv ed only the one application, and four w ere tre a te d again a t tw o-w eek in te rv als. T his schedule gave a total of five applications to the f ir s t group ranging down to one application in the la s t group. F ourteen plants of each v a rie ty constituted the controls. The re s u lts a re sum m arized in Table V. None of the controls of e ith e r v a rie ty showed any indication of seed stalk elongation a t the com pletion of the experim ent on May 7. In both v a rie tie s it appeared that four applications at tw o-w eek in te rv a ls sta rtin g when the plants had eight to nine leaves and ro o t Flowering Induced in Chantenay and Imperator Carrots Sprayed with 100 ppm Gibberellin in Aqueous Solution 33 3 CM 3 o 0 > 1-4 ■ ■M O 3 T3 CM O fn Q 0j CM CM CO CM o u3 O CO •w 3 0-4 3 rH O CM CM U o cd u O CD 04 CM vO CM VO lO vO CM CM On vO vO 00 O in in CM XI • 1-4 CD >> X 3a 3 CD cCDct £3 cd o UO t"C M CM vO 00 o CM X! U 4-1 < W ° s O cd 0 £ X ■ o 0 cd ■u 3 3 O. Q cn CO 1 vO ON I >> OO 4-> 0 u *»-*I CO 3 .3 UO 3 .3 XJ 0 Oh 3 i> CM U 0 X3 34 CO C3 31 1— a 0 Oh CO m >,.2 co in cd a* <3 3 w CNI OO i/O CM in co CM in co M O 0 X U . d ia m e te rs of 1.0 to 1. 5 cm induced the m ost rap id bolting. In both v a rie tie s the four p lan ts receiv in g th is trea tm e n t showed flow er buds by A pril 27 o r e a r lie r . Im p erato r responded m ore rapidly than Chantenay, and a ll the Im p era to r p lan ts had produced v isible flow er buds by A pril 27. Five appli­ cations sta rtin g when six to seven leaves w ere visible and the ro o ts w ere 0. 3 to 0. 5 cm in d iam eter also produced uniform flow ering. The response to o th er tre a tm e n ts was not uniform . Flow er induction from 0 to 50 p ercen t o ccu rred with only one appli­ cation. However, with only one application, the seedstalks w ere v ery sh o rt. Follow ing four o r five applications, the m ean plant height was 80 to 175 cm a t tim e of an thesis. Under these conditions it appeared possible to obtain re la tiv e ly uniform flow ering about seven to eight months a fte r seeding in the greenhouse, and about four to five m onths a fte r the fir s t application of gibberellin. The E ffect of P a rtia l Cold Induction Plus G ibberellin on Flow ering The experim ent was designed to determ ine if the combination of cold sto rag e and gibberellin hastened flow ering over e ith er trea tm e n t alone. M ature plants from seed planted O ctober 22, 1956 and grown in the greenhouse at a night te m p e ra tu re of 55° F w ere used. tenay, Im p erato r and D anvers. The v a rie tie s w ere Nantes, Chan­ F ou r plants w ere used for each gibberellin tre a tm e n t and four p lants w ere reta in e d as controls. All the plants w ere p laced in cold sto rag e at 38° F on M arch 12, 1957 fo r p erio d s of two, four and six w eeks. Some w ere left in pots under flu o rescen t lights. A second group receiv ed a single application of 100 ppm gibberellin before sto rag e. In a th ird group the plants w ere rem oved from the pots, topped and sto re d in damp e a rth to stim ulate roots from the field. At the tim e of storage the ro o ts w ere 1. 5 to 2. 0 cm in diam eter. Nine applications of gibberellin a t 100 ppm w ere applied at weekly in terv als, sta rtin g the day they w ere r e ­ m oved from cold sto rag e. All the tre a te d plants showed at le a st som e seed stalk elongation, a l­ though som e of them did not becom e reproductive. The data sum m arized in Table VI show that w ith the combination of cold and gibberellin all the Nantes and D anvers plan ts becam e reproductive. Flow ering was also ob­ se rv e d in one to th re e of the four check plants of Nantes and D anvers, which receiv ed cold trea tm e n t alone. It m ay be concluded that plants given six weeks cold trea tm e n t flow ered e a r lie r than those given two o r four w eeks. The cold induction alone up to six weeks was not enough to induce 100 p ercen t flow ering since th ere was no difference in the bolting of the checks following two, four o r six weeks of cold sto rag e. D uring m ost of the tim e following cold induction, the g ree n ­ house was over 70° F and often over 80° F. This m ay have re ta rd e d o r n u lli­ fied any useful benefit caused by the p a rtia l cold induction. Using gibberellin 36. s ■ — I O pH CM O O 1— I o o o O — H o o o Or HO O O O O O O I I I I I I O O O r - H O ’-H O O > — ■ I i 1 1 1 c o c o c o t—I CM CM co ^ CO '— I > — I i— I i— I O O T3 fl O U TJ ■ — t o l l 0 £+-> o "O o n ft H H CM CO ^ CO 03 0 oc; a -H O O O ’— I O O < — f c o<- HO •— I O O O i—t O ’■H »—t <—1 o o ON Xf CD 0 0 £ 'O X! O U i d0 U H pi g Q S O H CM ^ co O 'O C M ^N O I 0 ^ 0 CO CO o CM "M1 nO cm ^ t 1 n o T3 0 XS §s *2 g £ oS >«» O H xj 2 cd Q d cd d > P 0 *r *4 “H I CO 4-> o ft M 0 d £ CO 4-> o ft d co u 0 & cd Q 11——^( d T3 ^ ft It S i rd Cd u 0 l™"t I ( d U g, B & a0 xs s ft g a * <—I O O CM ^ no 0 d M X> O XI ■5b u ft ~ X, Q 0 •ft * d s0 IS .2 cd XI 0 7? ^R J -J u O c o '—I CM ^ 0 0 X ‘5b' O nO o d O d ft -r-t 03 oH 0d„ „d ■ cd •i_l ^ cd -d d -*-> O Number of plants reproductive out of a total of four plants receiving treatment. Flower Induction in Carrots Subjected to Partial Cold Induction at 38° F plus Gibberellin CM CO 3. % X3 p lu s cold re s u lte d in no saving in tim e as about 140 days elapsed before an th esis, which was about the sam e as with the conventional cold induc­ tio n m ethod alone. The use of G ibberellin to Induce Flow ering in the F ield The p revious two experim ents had shown that bolting and flow ering could be induced in the greenhouse, w here th ere was no lim it to the grow ­ ing season. At the Muck E xperim ental F arm two stra in s of Chantenay w ere seeded adjacent to each o th er in 10-foot row s on A pril 20, 1957. One of th ese stra in s TBf had been previously used fo r all w ork with Chantenay and had shown little o r no indication of containing seed which would produce annual p lan ts. The oth er line TA' was Chantenay seed of a line which had been o bserved to produce 20 p e rc en t field bolting in 1956. A sim ila r ex­ p erim en t was planted in 20-foot row s on A pril 22 at M oreland Muck F a rm s n e a r G rant, M ichigan. Because of a shortage of seed a non-bolting line of Im p erato r w as used at G rant instead of line *Af. T his was the sam e line used the p revious w inter in the greenhouse. C oncentrations of 100 ppm and 1, 000 ppm of gibberellin in aqueous solution w ere applied on plants ranging from the two to fourteen leaf stage at the tim e of f ir s t application. The num ber of applications ranged from one to eleven. Table VII su m m arizes the observations from th is experim ent. In all c a se s, it appeared that for e a rlie s t flow ering the f ir s t trea tm e n t should be Plant Height and Bolting in Carrots on September 10, 1957 Following Treatment with G ibberellin 38. d 0 bo d o o o o o o o o o o in \o o m o o r - O n vo r— o © o in CM oIs- oo o o o co in m t O o l O m o o o o o r*- o i-H CM 1 i-H o r- o m i—i m o o o o r-H o o © o 0 XI o o %S2 uo 4-> cd u0 a o to cm o in && co 3 s §I & a I— m D-i -H s m o o s ■§> cd -T-i CD 0 leS s0 ® bo £0 *» o -S H -H £Q & <2 (S cd a l— l CO CO CD o m o o -W [ > o o a CO o © bo d a o bo d £0 2O d 0 oCM o^ o r~- or- o o o VO vo P-. © >n cd +-> d CD CD •M XI XS a0 u o CO tdn PL, 0 I, 0 x o CO r O <—f o 00 1 o 1—t o o CO rH 1 o o 1—1 o i r- ON t od4 o •^4-> d cd o u U d o O o o 1—1 o 1—1 0 o o o o 1—1 o 1— 1 o o o © r—l o 1—1 I— ( u 00 1“ 1 4-J pH C0O co H 0 £R 0 H X I 0 M S ^ 0 £ O H d CM I co r-i m i o o n d o O O when the plants a re at the six to eight leaf stage, and the ro o ts a re ju st s ta rtin g to sw ell. C oncentrations of 1, 000 ppm gibberellin in c o n trast to 100 ppm re su lte d in ta lle r plants with th ic k er stem s. Five o r six weekly applications produced the e a rlie s t flow ering. With four tre a tm e n ts, all Im p erato r plants had developed flower buds by Septem ber 10, but w ere la te r than with five o r six trea tm e n ts. L ess than four applications generally induced only a sm all proportion of the p lan ts to becom e reproductive, and usually only on short seedstalks. Ten o r eleven applications at 1, 000 ppm sta rtin g at the two and four leaf stage, re su lte d in v e ry bushy and d istorted p lants. A ll the plants flow er­ ed but w ere no e a r lie r than those receiv in g five o r six trea tm e n ts sta rtin g at the six to eight leaf stage. In a ll c ase s, in resp o n se to gibberellin the bolting line A tended to bolt much m ore re a d ily than the non-bolting line B. Im p erato r bolted m ore rea d ily than line B in resp o n se to gibberellin, although it had no g re a te r n a tu ra l tendency to bolt. This was consistent with observations in the g ree n ­ house during the previous w inter. When g ibberellin tre a tm e n ts w ere sta rte d at the six leaf stage, th e re w as no difference in tim e req u ire d to flow er following bi-weekly, as com pared w ith w eekly applications, provided the num ber of applications was the sam e. T he E ffect of G ibberellin on F ield Induction of F low ering in C a rro t Plants D iffering in Bolting Tendency E xperience in 1956 and 1957 indicated that following a sufficient p eriod of cold induction som e plants in m ost v a rie tie s can be induced to bolt natu rally in the field. Five to seven applications of gibberellin at 100 o r 1, 000 ppm cau sed m o st of the tre a te d plants to bolt and becom e reproductive by late fall. P a rtic u la rly w ith the natu rally slow bolting stra in s this percentage was m uch higher than w ith cold induction alone. A rep lica ted experim ent was perfo rm ed in the sum m er of 1958 using the sam e non-bolting Chantenay line B as in 1957. F o r the bolting v a rie ty a C hantenay stra in w as used which had bolted n aturally 20 p e rc en t in 1957. The bolting and non-bolting stra in s w ere planted a ltern ately in 10-foot row s. T hree rep licatio n s w ere used with one, th ree and five applications of 100 and 1, 000 ppm gibberellin at weekly in terv als sta rtin g on June 23 when the plants w ere at the six to seven leaf stage. The seed w as planted at the Muck E xperim ental F arm on May 12, when little natural field induction w as expected to occur. b e r 13. The b o lters and non-bolters w ere counted on Septem ­ The percentage of n atu ral flow ering b o lters and of those with elongated ste m s of over 18 inches following the various treatm en ts, a re rec o rd ed in Table VIII. One application of gibberellin had no effect at concentrations of 100 or 41 03 0 O ■0g Stem Development In Carrots Following Treatment with Gibberellin o r(D 5m bO 0) 0 2 I■■■I o <2 • o ^ • * o '—I O • vo \0 CO I— I v o OO i> oo CO ON lO • • cm • 1-4 aj U 0 4-> bo 0 aj rM • 2 m I—« •“ I ON CM CM bo 0 8 o11 i« TJ CL) O a0 o uD C ei 0 a 0 o U a 04 TO 0 M aj 0-i T3 o • CM >-M o• o• o o mo • * o o I1 1 OO o oIO CM 1 o o 0 O 0 o o o-M O o o o o--M oo o 5M 0 X) a0 2: 0 4—> 0 O •1-4 i 1 04 04 < co co io uo •i-i M o a 0 0 TO bo 0 -I-, > aj aj 0 TO oa O •i-t TO 03 8) 3 5-t TD TO TO + -> xi 5-1 <+-, o o o’-H oo TO so 0 M—1 0 aj Pi 05M =1 TO aj 0 a a 0 • " -j 5M •H X3 1 O 10 00 aj 0 8* 04 TO 0 04 5m 0 X3 B 0 0 0 O T) 0 TO £ . 1, 000 ppm . T h ree applications at 100 ppm produced no effects. At 1, 000 ppm 3 .4 p e rc en t of the p lants in the bolting line produced seedstalks as a re s u lt of the gibberellin trea tm e n t (Table VIH). These could e a sily be distinguished from the n atu ral b o lte rs by growth c h a ra c te ristic s . The to tal num ber of n a tu ra l b o lte rs plus those elongated as a re s u lt of gibberellin was not h igher than the 2 2 .4 p e rc e n t o ccu rrin g in the check. Following five applications of gibberellin at 100 ppm, 10. 9 p ercen t of the p lan ts of the non-bolting line produced stem s of 18 inches o r m ore, and a ll the rem aining plants had stem s of 8 to 12 inches with the typical gibberellin induced appearance. p e rc e n t b o lte rs. The untreated plants produced only 0. 8 The five applications of 1, 000 ppm also produced 100 p e rc e n t stem elongation w ith 34.4 p e rc en t over 18 inches on Septem ber 13 (Table VIII). In the two lines used in this experim ent gibberellin could not be con­ sid e re d a re lia b le m eans to elim inate plants with a bolting tendency. With five applications the non-bolting plants produced stem s of 8 to 12 inches and could not be sto re d to produce seed la te r. T hree applications could not have been used as a selective m echanism in 1958 because the unusually cold sp rin g re su lte d in m ore n atu ral bolting than usual. In a w a rm e r y e ar when they would not have bolted naturally, some plants with a n atu ral bolting te n ­ dency m ight have been induced to bolt by one o r th ree applications of gibberellin. G ib b erellin as a M eans of Selection fo r Slow Bolting in the Field In the sum m er of 1957, 112 experim ental hybrid c a rro t lines w ere being grown a s p a r t of the c a rro t breeding p ro g ram at the Muck E xperim ental F a rm . In each 12-foot row, the f ir s t th ree feet w ere sprayed at weekly in terv als with g ib b erellin at 1, 000 ppm fo r a p erio d of six w eeks. Spraying w as sta rte d on June 14 when m ost p lan ts had reach ed th e ir six leaf stage. The rem ain d er of the row serv ed a s a control fo r determ ining the natural annual flow ering in each line. Since the c a rro ts w ere seeded on May 1, they received a r e la ­ tiv ely long p erio d of n atu ral cold induction in the spring. The re s u lts sum m arized in Table IX indicate that generally the lines with a n atu ral tendency to bolt also showed the g re a te st resp o n se to g ibber­ ellin . However, of the 38 hybrids th e re w ere nine exceptions to this general ru le . T hese nine lines produced elongated seedstalks as a re s u lt of g ib b er­ ellin, but did not produce any natural b o lters. M ost of the 64 lines which did not respond to gibberellin also produced no n atural b o lters. The th ree notable exceptions to th is general relationship w ere lines 246, 261 and 262 (Table IX), which produced a rela tiv e ly proportion of b o lte rs. shows a line which responded vigorously to gibberellin. Figure 3 TABLE IX The R esponse of D ifference H ybrid C a rro t L ines to T reatm en t with G ibberellin in Com ­ p a riso n w ith th e ir N atural Bolting Tendency Hybrid Line No. G ib b er­ No. of e llin R e­ N atural sponse Bolters + > - 1/ H ybrid G ibber­ No. of ellin R e­ N atural Line No. sponse Bolters +, - Hybrid Line No. G ibber­ No. of ellin R e­ N atural Bolters sponse +, - 101 0 0 195 + 0* 269 0 2 273 + 103 195 A 6 0 103A + 0* 275 1 + 195B 277 0 197 + 0* 105 0 + 15 + 0 197A 0* 279 107 4 281 + 4 0 199 107A 1 283 + 199A 0 19 107B 3 285 + 201 2 5 109 4 287 8 203 0 + 109A 2 + 207 0 289 0 115 0 0 313 207A 0 115B 0 8 315 209 1 117 317 + 0* + 2 209A 3 117A + 0* 7 319 + 221 41 + 117B 0 415 1 225 3 117C 0 417 0 225A 11 + 117D + 0* 14 419 + 231 0 119 0* + 421 2 240 0 121 431 + 10 4 241 0 123 1 433 0 243 0 123A + 0* 435 0 245 0 125 + 2 437 10* 246 0 127 + 1 0 439 249 0 129 441 9 6 + + 251 0 133 443 0 3 + 252 6 + 179 0 445 2 254 16 + 181A + 460 0* 0 259 46 + 183 462 0 0 260 0 185 + 17* 474 1 261 0 185A 0 480 11* 262 6 + 185C 484 0 0 264 2 + 185D 487 0 0 265 4 + 187 493 0 0 267 0 189 497 0 0 269 16 + 183A 38 lin es responded to gibberellin and 9 of these had no n atu ral b o lte rs. 74 lines did not respond to gibberellin, 16 of these had som e n atu ral b o lters, but over 50% of b o lte rs in the 74 lines which did not respond to gibberellin w ere in 3 lines. — Me ans a ll of the plan ts responded to gibberellin and w ere reproductive by Septem ber 15. - Indicates lim ited stem elongation with no flower bud development by Septem ber 15. * The num ber of n a tu ra l b o lte rs shows no positive relationship to resp o n se to gibberellin. ). F ig u re 3. R esponse to gibberellin in a c a rro t hybrid following six weekly applications of gibberellin at 1, 000 ppm beginning at the 6 leaf stage. Right, tre a te d plants; left, untreated plants in the sam e row. The resp o n se to gibberellin w as atypical a s m ost lin es responding so vigorously to gibberellin also produced som e n atu ral b o lte rs. 45. Inheritance of Annual V ersus Biennial C h a rac te r in C a rro ts Bolting p lants selected in 1956 from foreign plant introductions, com ­ m e rc ia l stra in s and breeding lines w ere c ro sse d with non-bolting plants. The b reed in g behaviour of these non-bolting plants was not known com pletely, but it w as hoped that som e of them m ight be homozygous. In o rd e r to m ake the c ro s se s, the cen ter um bellets and cen ter flo rets of the m o re m atu re o u ter um bellets w ere rem oved from one o r two um bels. The introduction of the pollen p a re n t um bels into the cages was delayed until all the a n th e rs had dehisced on the fem ale plant. The cut stem of the pollen p a re n t was placed in w ater to keep it alive long enough for pollination. The p a re n t um bels w ere p ro tected with a cloth cage and flies w ere introduced to tra n s fe r the pollen. D uring sum m er 1956 one line (No. 165) was noted to have 31 p ercen t b o lte rs. In Novem ber, rem nant seed of line 165 was planted in the g reen ­ house and bolting seg reg ates w ere c ro sse d with th ree plants from non-bolting lines, using the sam e m ethod of hybridization as was used in the field in 1956. O ther annual plants selected in the field w ere transplanted to the greenhouse and used a s p a re n ts in com bination w ith annual and biennial plants. Table X tabulates the c ro s s e s made in sum m er 1956 and w inter 1956-57, and the bolting ra tio s observed in sum m ers 1957 and 1958. Seed from all the c ro s s e s m ade in sum m er 1956 and w inter 1956-57 46. TABLE X Segregation fo r Bolting O bserved in and Progenies of Bolting and Non-bolting P a ren ts in 1957 and 1958 P a re n t P lants and F lo w erin g Habit N on-bolting X bolting 1445 X 368 1440 X 368 1440 X 378 1412 X 437 1462 X 2001-1 2395 X 165-2 2165 X 165-2 2208 X 2001-2 2638 X 2001-2 2638 X 743-1 Bolting selfed 368 378 165-2 165-6 743 2001-2 N on-bolting selfed 1440 2395 2165 2638 1154 1166 1192 1226 Phenotype in 1957 Bolting Non-bolting 43 18 40 25 35 31 118 72 6 0 36 79 14 7 60 0 0 0 30 23 7 65 40 7 86 123 - 23 44 - 4 25 - 20 0 9 - 23 - 0 - 0 5 0 15 9 6 0 20 0 100 0 100 45 85 19 32 45 38 - - - 0 20 1 - - 5 12 13 0 140 50 0 Phenotype in 1958^ Bolting Non-bolting 1 - - 0 12 5 22 1 11 - — N on-bolting X non-bolting 1189 1149 1159 1189 1193 1219 X X X X X X 1154 1166 1166 1192 1192 1226 37 41 18 55 25 2 *U sing rem nant seed in 1958. 80 16 64 42 36 100 15 16 38 1 24 4 10 0 1 17 w as planted in the field in 1957. M ost of the seed from c ro s s e s made in the w in ter w ere planted too late to receiv e a good cold induction in the field. Roots of non-bolting plants from lines grown in the field in 1957 w ere r e ­ m oved to the greenhouse in the fall. / In the sp rin g of 1958 w eather conditions w ere v ery unusual with a com bination of drought and extrem e of tem p eratu re. T his combination r e ­ sulted in p oor stands of c a rro ts in m any of the 235 lines planted on A pril 15 fo r genetic an aly sis. With th is e a rly planting and a long cool season it was expected that alm o st all plants with a tendency to bolt would do so. Incom ­ p lete p en etran ce of genes fo r bolting w ere expected to be r a r e . It is difficult to determ ine inheritance of the bolting habit because of the g re a t influence of environm ent on expression of the flow ering c h a ra c te r­ istic . It can be noted in Table X that certain progenies showed m arked dif­ fere n ce s in segregation in two different y e ars. One of the m ost strik in g exam ples of th is environm ental influence is the progeny 1445 X 368 which seg reg ated 43 b o lters to 31 non-bolters in 1957, and 13 b o lters to 1 n o n -b o lter in 1958. Obviously any genetic data to be conclusive would have to be secu red under a c arefu lly controlled environm ent. T h ere is some evidence in the data from two different y ears that c e r ­ tain lin es a re ra th e r easy to bolt and can be expected to behave as annuals even in y e a rs when they a re not subjected to prolonged cold exposure in the 48. sp rin g . An exam ple of such a line is No. 743 (Table X) which bolted alm ost a s m uch in 1957 w ith little induction as it did in 1958 w ith plenty of cold ex ­ p o su re. The data in Table XI show a wide range of segregation in progenies from b o lte rs and n o n -b o lters. T his lack of a c le a r cut p a tte rn of segregation to g eth er w ith the m any degrees of bolting expression suggest that inheritance is quantitative with an undeterm ined num ber of genes operating. T h ere is a tendency tow ard dominance of the annual habit indicated in se v e ra l pro g enies liste d in Table XI. which non-bolting The four back c ro ss progenies in plants w ere c ro sse d to b o lters all segregated b o lters 100 p ercen t, except one (444 X 39) which produced only one non-bolter in 50 p lan ts. M ost of the F 2 progenies from non-bolting F^ plants segregated a ra th e r high p roportion of b o lte rs. Unfortunately, the non-bolting se g re ­ gates in F j w ere selected in 1957, a year of inadequate induction, and could e a sily have been plants c a rry in g genes for bolting. In Table XU a re liste d F progenies from five c ro s se s. Again the F t 2 plants w ere selected for flow ering habit under conditions of inadequate induc­ tion in 1957. However, som e p ro g re s s toward selection for non-bolting can be observed. T here a re generally few er bolters in the progenies of non-bolt­ ing plants than in those of bolting plants. It should be noted also that the e a rly bolting F^ plants produced a m uch higher proportion of b o lters in the F^ TABLE XI Segregation fo r Bolting Habit in C a rro ts Plant No. 450 450 453 453 475 439 439 647 439 449 449 444 444 448 448 X 17 X 17 X 54 X 647 X 33 X 39 X 639 484 X 13 481 X 18 18 482 X 635 480 483 489 490 Pedigree Phenotype of P arent Plant 2165-1 self 165-2 self 2165-1 X 165-2 (2165-1 X 165-2) 165-2-1 (2165-1 X 165-2) (2165-1 X 165-2) 165-2-1 (2165-1 X 165-2) 743 self (2165-1 X 165-2) 743-1 (2165-1 X 165-2)2165-1-1 2165-1-1 (2165-1 X 165-2) (2165-1 X 165-2) 165-6-1 (2165-1 X 165-2) (2165 X 165-2) 165-6-2 (2165-1 X 165-2) (2165-1 X 165-2)2165-1-2 (2165-1 X 165-2) NB B NB X B NB X B NB NB X B NB B NB X B NB X NB NB NB NB X B NB NB X B NB NB X NB NB (1159 X 1166) (1159 X 1166) 165-2-3 (1159 X 1166) 155-2-4 165-2-4 (1159 X 1166) 2638-1 (1159 X 1166) (1159 X 1166) (1159 X 1166) (1159 X 1166) NBX NB X NB X B NBX NB NB NB B NB B B NB G en er­ ation Pj F1 Phenotype 1957 1958 NB NB B _B 1 10 0 5 0 79 20 9 0 100 7 36 BC F2 F2 Pi BC BC S2 F2 BC BC F2 F2 F2 F2 18 64 0 20 14 8 0 5 1 8 20 8 14 2 0 7 1 14 13 27 21 8 16 29 41 13 13 14 28 3 13 24 22 F2 F2 33 4 33 37 49 f 2 BC BC BC P, BC 0 10 31 BC F1 20 0 15 2 28 25 14 59 3 50. d .)_ co o •r-t > d oCO a « d rJ Oh CD U d W o CO e g T3 CD M — I i— l 0 co CO d d fe T> CD in n bO co m ON CO O d cq NO c8 CD •*-> o d m t"- (M CM hco NO in CO co cm o ON NO CO XJ 0 a rd p-i 0 CO CQ H 2: CM IX. O o CM CM m On NO in CM T*l O bo d CD CD CO 0 .Q 1 d B o £ CQ co CM OO CM CM CO in co o CM 2; £Q M — I o CD & 4-> B CD p-l r-- m ON CQ 0 oCO NO r- in CM NO oo in in m NO NO cm ON cm ON CO NO ft 4-J O d JB Ph CQ <— I co CM [L| co CL, -UJ d 0 ah O NO NO X X X X X ON OO ON ON 00 CO ON ON in o n than did the plan ts c la ssified as late b o lters. Selection would be m ore effective if p a re n t plan ts w ere exposed to longer induction p erio d s than was the c ase in 1957. E xperience with a large num ber of inbred lines shows th at selectio n fo r non-bolting combined w ith inbreeding can alm ost elim inate the bolting tendency. The ease of establishing such lines indicates that th e re a re rela tiv e ly few genes involved. DISCUSSION AND CONCLUSIONS Flow er Induction by Cold T reatm ent and V ernalization T h ere w ere distinct v a rie ta l differences in c a rro ts as to the duration of cold exposure re q u ire d to induce flowering. Nantes req u ire d a sh o rte r p erio d of cold exposure than Chantenay or Im p erato r. Sakr and Thompson’s (1942) w ork showed that the v a rie ty F ren ch F orcing req u ired a sh o rte r p e r ­ iod than N antes. However, stra in s within the v a rie tie s v aried ju st as widely in re q u ire m e n ts as the v a rie tie s them selves. A 10-week storage period seem ed to be optimum for uniform bolting of all v a rie tie s within the sh o rtest period following fall h arv est. Ten weeks cold induction was not always adequate for a few lines o r individual plants in c a rro t v a rie ty im provem ent studies, and these plants would be highly d esirab le as a source of genes with re sista n c e to bolting. age p e rio d s longer than 10 With s to r ­ weeks, th ere is not enough tim e for seed to m ature and rip en fo r field planting in late A pril to secu re a satisfac to ry bolting te st. F u rth e r, with the longer storage period, it is im possible to elim inate plants that m ay bolt v e ry soon a fte r planting. re a lly late field b o lte rs. These e a rly greenhouse b o lters a re This failu re to select against bolting also occurs when ro o ts a re sto re d over the w inter and grown in the field the following year. Follow ing 10 weeks of cold induction at 40° F, any plant showing seedstalk elongation in the greenhouse w ithin the fir s t two to th ree weeks a fte r p lan t­ ing should be d isre g ard ed as having a n atural bolting tendency. R esponse to cold exposure seem s to be quite independent of root size. Even the ro o ts h a rv e ste d at 8 weeks of age and only 1. 2 cm in diam eter p ro ­ duced seed stalk s and flow ers following cold induction. Few er of the sm all ro o ts bolted and they w ere th re e weeks la te r than larg e m ature ro o ts. The delay in flow ering appeared to re s u lt from a lack of vigour o r re s e rv e in the sm all ro o ts, ra th e r than th e ir being m ore difficult to induce. This is in c o n tra st to onions, which according to Thompson and Smith (1938), m ust re a c h a c e rta in size before they can be induced to flow er. ing to Boswell (1929), cabbage stem s m ust re a c h a 6 Likewise, a c c o rd ­ mm diam eter before the plant can be induced to bolt. Paw ar and Thompson (1950) observed that in c e le ry the older p lants at the tim e of cold treatm en t went to seed fa s te r than young p lan ts. However, the total tim e from seed to seed was sh o rtest when tw o-m onth-old plants w ere subjected to cold induction. They found that plants given cold induction at any age would bolt eventually, and that age ra th e r than size w as the im portant facto r. Following the s ta rt of seed stalk elongation in the greenhouse at 55° F, Dickson (1956) suggested that ra is in g the tem p eratu re to 70° F would hasten a n th esis and seed m atu rity . In a tria l to te s t th is assum ption, the interval betw een the f ir s t appearance of the seedstalk and anthesis was reduced by two to th re e weeks when the plants w ere moved to the 70° F house at the f ir s t v isib le indication of bolting. Also, the plants w ere a foot sh o rte r than when grown at 55° F. T his saving of two to th ree weeks is of consid­ e ra b le im portance because it m ay p e rm it a greenhouse seed crop to be re a d y fo r planting e a rly in spring. This ra is in g of the tem p era tu re was in line w ith Thom pson’s (1929) suggestion for celery. V ernalization of c a rro t seed does not appear to be very successful as a m ethod of inducing bolting. Some bolting was induced, but the p e r ­ centage was not as high as that resu ltin g from e a rly planting in the field. The m o re successful induction of young seedlings in the field com pared with v ern alizatio n is in agreem ent with Chesnokov’s (1932) statem ent that young seedlings ra th e r than slightly germ inated seed gave a higher p e rc e n ­ tage of se e d e rs following 50 days of cold treatm en t. E a rly sp rin g planting appears to be a v ery satisfac to ry method to select fo r non-bolting if th ere is a long cool spring. In the tria l in 1957, bolting was d irec tly proportional to the e a rlin e ss of planting. p ecially tru e of lines which inherently produce field b o lters. This was e s ­ T here was little o r no effect of e a rly planting on lines that w ere inherently non-bolting. L eav­ ing the p lan ts in the field sufficiently late in the fall to allow tim e for bolting is ju st as im portant as e a rly planting. By harvesting in e a rly August, ra th e r than in e a rly Septem ber, lines m ight be selected for potential seed production w hich a t the la te r h a rv e st would be reje cted on the b asis of bolting tendency. O bservations from the tr ia l in 1957 (Table III) showed that th e re w ere two to th re e tim es a s m any b o lte rs visible on Septem ber 4 as on August 7. If e a rly h a rv e st is p ra c tic e d in o rd e r to avoid la rg e roots and excessive top injury due to d isease, then v ery rig id culling for e a rly bolting in the greenhouse is n c e ssa ry . T his is in o rd e r to avoid actual selection for, ra th e r than against e a rly bolting. An attem pt w as m ade to estim ate a possible minimum num ber of hours below a given te m p era tu re which would induce bolting in all plants with any n a tu ra l tendency to bolt. A p re c is e m easu re of cold exposure needed for in ­ duction would provide the b re e d e r with a method to evaluate his selections fo r bolting re sista n c e . G row ers and p ro c e ss o rs m ight use such data to fo re ­ c a st if heavy bolting is to be expected in v a rie tie s with a tendency to bolt. The a r b itr a r y base te m p era tu re above which no induction of bolting o ccu rred was set at 50° F. In 1957, no definite tim e lim it appeared to be reached, a l­ though the plants which bolted f ir s t w ere exposed to 600 hours of tem p eratu res below 50° F. In 1958, a th resh o ld appeared to be reached at approxim ately 650 hours below 50° F at the soil surface. T here was no fu rth er in c re a se in bolting with 800 hours exposure, and at 500 hours th ere was significantly le s s bolting than th ere was with 650 hours. It is suggested that 600 to 700 hours below 50° F is re q u ire d to insure induction of all potential b o lte rs. is n e c e ssa ry in o rd e r to estab lish m ore p re c ise lim its. F u rth e r w ork The soil su rface te m p e ra tu re appears to be m ore reliab le because it clo sely p a ra lle ls the tem p era tu re a t the crown of the plant. T em p eratu res above the p lants v a ry considerably from that at the soil surface, the d iffer­ ence depending on soil type and m o isture content. Induction of Flow ering in C a rro ts by G ibberellin Lang (1957) and W ittwer and Bukovac (1956, 1957) rep o rted that som e c a r r o ts could be m ade to flow er following the application of gibberellin. However, th ere was a lack of uniform ity in the tim e req u ire d for flowering. In the f ir s t greenhouse experim ent with Chantenay and Im perator, a v a rie ta l difference in resp o n se to gibberellin was evident with Im perator r e ­ sponding e a rlie r and m ore uniform ly than Chantenay. N either stra in showed any tendency to bolt in the field even when planted e a rly in the spring. In the greenhouse, flow ering was relativ ely uniform following four o r five tre a tm e n ts with gibberellin, but not following one to th ree trea tm e n ts. In both stra in s , the highest percentage of bolting was induced when the plants w ere a t the 7-8 leaf stage at the f ir s t treatm ent. In the f irs t greenhouse e x ­ perim en t, the te m p era tu re during m ost of the life of the plants was around 60° F, which is five to ten degrees above the threshold for natural induction. It ap p ears from th is and other investigations that biennials respond to gibberellin s b est at a tem p era tu re approaching, but somewhat higher than the threshold te m p e ra tu re for induction by cold. Because of lim ited space the num ber of plants in the greenhouse e x p e ri­ m ent w as too sm all to avoid the variation caused by th e ir biological d iffer­ ences. F ortunately, the differences observed w ere g rea t enough that con­ clusions could be draw n from the sm all populations used. In the second experim ent using m ature roots, the re s u lts w ere much m ore v a riab le. It ap p ears th ere is an optimum stage of m atu rity for the f ir s t t r e a t ­ m ent, that w ill produce re la tiv e ly uniform flowering. This also appeared to be tru e in the field w here m ost of the hybrid lines w ere fairly uniform in th e ir resp o n se to six weekly gibberellin applications. When plants w ere firs t sp ray ed at a stage la te r than 6-7 leaves, the response v aried widely, r e ­ sem bling that of p lants grown from m ature roots in the greenhouse. This was sim ila r to W ittwer and Bukovac's (1957) re s u lts with celery. Combining p a rtia l cold treatm en t with gibberellin re su lte d in a saving of labor p ro p o rtional to the tim e they w ere in the cold storage. C arro t roots subjected to ten weeks cold storage alone, o r to p a rtia l cold induction (two to six weeks) plus gibberellin o r to gibberellin alone req u ired a p eriod of approx­ im ately 140 days to anthesis from the beginning of treatm en t. The plants tre a te d w ith gibberellin in m any case s appeared to produce norm al flow ers a t the com m encem ent of anthesis. However, a fte r being caged for two to th re e w eeks with flie s fo r pollenators, little o r no seed had developed and the ovules appeared to be v ery much sm aller than on untreated plants. The ir re g u la r bolting following treatm en t with gibberellin alone m akes its use for seed production in the breeding program im practical. Some vegetative and flow ering resp o n se s to gibberellin a re illu stra te d in F ig u re 4. However, if adequate cold treatm en t (10 weeks o r m ore) is given and som e p lan ts a re still v e ry slow to produce seedstalks, a few applications of g ib b erellin can be used to encourage flowering. According to Honma (1957) th is w as sim ila r to the resp o n se of som e lines of celery which w ere induced to flow er by gibberellin trea tm e n t a fte r they failed to bolt following cold induction. The m ost useful application fo r gibberellin on c a rro ts m ay be to select lin es which have a tendency to bolt but which a re not bolting because of eith er late planting o r a w arm spring. The use of gibberellin in this way might p e r ­ m it h a rv e st in August instead of Septem ber, while still allowing adequate selection against bolting. T h ere ap p ears to be definite genetic differences between lines in th eir resp o n se to gibberellin. Those with a n atural tendency to bolt will produce seed stalk s readily, and those without any such n atural tendency will usually produce seed stalk s v ery slowly in response to gibberellin. T herefore, g ib b er­ ellin m ay act as a stim ulus to bolting provided the genes for flow er develop­ m ent a re subsequently activated. However, these with a strong biennial habit (no b o lters) re q u ire cold treatm en t to activate genes for flow er form ation. F ig u re 4 R esponse to gibberellin of greenhouse grown c a rro ts . , 2, 3, 4 and 5 a re v ario u s stages of flow er developm ent at the end of 1 4 to 5 foot stem s, following tre a tm e n t w ith gibberellin. un treated um bel. 6 is a norm al 7, com paring gibberellin tre a te d and a cold induced plant at the sam e stage of flow er developm ent. cold induced fo r 10 weeks and then planted. The sm a lle r plant was The g ibberellin tre a te d plant was potted at the sam e tim e the o th er plant went into cold s to r ­ age. Photographs w ere taken four weeks a fte r em ergence from cold o r 14 weeks a fte r g ibberellin tre a te d ro o ts w ere planted in the greenhouse. 60. In such plants gibberellin alone seem s to only cause stem elongation without n o rm al flow er form ation. G ibberellin usually caused stem elongation and flow er bud development on a stem often two and one-half tim es the norm al height. The flow er buds w ere usually norm al if the stem elongation was not excessive. w as over 120 When the stem cm, o r when the bud was slow to develop, it was often abnorm al. In the greenhouse in w inter p a rtia l m ale ste rility often developed along w ith a g en eral reduction in flow er vigor. This m ay be the re s u lt of excessive vege­ tativ e growth a t the expense of norm al reproductive development. P erhaps gibberellin induces elongation m ore than flow er initiation; and following stem elongation, the genes fo r flow ering m ay become activated eventually. Thus, if the plants w ere heterozygous for non-bolting, they might flow er m ore rea d ily than if homozygous for non-bolting. This may account fo r v ariatio n observed following gibberellin treatm ent, since bolting is not controlled by a single gene. This m ay explain why plants grown near th e ir th resh o ld te m p era tu re fo r bolting induction respond to gibberellin m ore read ily than at higher te m p e ra tu re s. Inheritance of Annual V ersus Biennial C h aracter in C a rro ts The re s u lts of inheritance studies indicate that bolting in c a rro ts m ust be influenced by se v era l genes. However, there is fairly conclusive evidence th at annual flow ering is dominant over biennial flowering. This is evident in se v e ra l back c ro s s e s of non-bolting F^ plants to the bolting p a re n t in which the b ack cro ss progeny w ere all b o lte rs. F u rth e r evidence that the annual tendency is dominant m ay be found in the fact that it has been r e la ­ tiv ely e asy to e stab lish non-bolting lines by inbreeding. F ro m the experim ents with segregating progenies planted in two dif­ fe re n t y e a rs it is c le a r that c ertain heterozygous plants m ay behave a s annuals under one set of conditions and as biennials under another. This type of b e ­ haviour suggests that incom plete dominance is involved. In a to tal of 235 lines segregating for flow ering habit some progenies produced ra tio s which fit a hypothesis in which bolting was assum ed to r e ­ su lt from the action of two m ajo r genes com plem entary in action and with one o r both being only p a rtia lly dominant. A ccording to this hypothesis the d iffer­ ent ra tio s o bserved in different y e ars could be accounted for by incom plete p en etran ce under conditions of inadequate induction such as o ccu rred in 1957. However, m any of the progenies failed to fit any of the th eoretical ra tio s which would re s u lt under th is hypothesis. F u rth erm o re, because of the un­ c e rta in cla ssifica tio n of p aren t m a te ria l in 1957 and the heavy lo ss of plants in 1958, the data available m ust be considered inadequate to determ ine a p re c is e inheritance p attern . In m any c ase s plants c la ssified as non-bolters m ay be slow bolting plants, heterozygous for one o r m ore genes with the bolting c h a ra c te ristic unexpressed because of insufficient cold exposure o r a shortened growing season. This is evident in the low er incidence of bolting in progenies from non-bolting F^ plants com pared with those from bolting F^ plants. The strength of the bolting tendency appears to be indicated by how quickly the p lan ts produce seedstalks a fte r spring induction. The e a rly bolting plants in F , produced a higher percentage of b o lters in F than did the F 1 plants which ^ 2t i bolted late in the fall. F o r the b re e d e r to develop tru e non-bolting lines for Michigan conditions he m ust plant e a rly to allow sufficient cold exposure for flow er induction. m u st also be allowed fo r the b o lters to develop p rio r to h arv est. Tim e In field ex­ p e rim e n ts in 1957 and 1958, the tim e of exposure to tem p era tu re s below 50° F to in su re elim ination of incipient b o lters was som ething in excess of 600 hours (Table IV). The length of growing season to allow b olters to develop should exceed 140 days. It should be possible to develop from plants of an adequately tested line an^ open pollinated v a rie ty in which no b o lters would develop even in the m ost ex trem e w eather conditions experienced in Michigan. SUMMARY V arie tal differences w ere found in the duration of cold req u ire d to induce bolting in c a rro ts . Of th re e v a rie tie s tested, Nantes req u ire d the sh o rte st tim e to bolt while Im perator w as slow est and Chantenay was in te r­ m ediate. All th ree v a rie tie s produced 95 to 100 p ercen t seedstalks in the sh o rte st tim e a fte r planting if given 10 weeks of induction by storage of m atu re ro o ts at about 40° F. Root size was found to have no m ajor effect on e ase of flow er induction. However, v ery sm all roots of 1 cm o r le ss in d iam eter produced seedstalks m ore slowly than larg e ro o ts. To produce anthesis as e a rly as possible the optimum growing te m ­ p e ra tu re s in the greenhouse following cold induction w ere 50 to 60° F until seed stalk s becam e visible, and 70° F until anthesis. By following this p ro ­ cedure two to th ree weeks w as saved in the tim e from planting to anthesis. F o r selection of non-bolting plants in the field adequate cold induction in the sp rin g is esse n tia l. A total of approxim ately 650 hours was found to be the m inim um p eriod of exposure to tem p era tu re s below 50° F req u ired to obtain seed stalk development in all plants with genetic tendency to bolt. Tim e m ust also be allowed for all bo lters to develop. be m o re im portant if sp rin g tem p era tu re s a re high. b o lte rs develop slowly. This was found to In a warm spring A p erio d of 140 days between seeding and h a rv e st allowed tim e for alm ost all potential b o lters to develop. Reducing the above p erio d by 15 days resu lte d in 5 to 10 percent reduction in the num ­ b er of b o lte rs in a heavily bolting line. In apparently non-bolting lines a few b o lte rs som etim es develop, if given the e x tra days in the fall. If th ese late b o lters a re not allowed to develop, and the ro o ts a re harvested and used in breeding p ro g ram s, they then become a source of genes for bolting. G ibberellin will produce bolting, especially in lines which have a n atu ral tendency to bolt. It appears to stim ulate e a rlie r bolting in lines which in w arm y e a rs would otherw ise segregate bo lters and non-bolter s. In this resp ec t, it m ay be of use a s an agent to select for non-bolting. As a m eans of acc elera tin g bolting in the greenhouse for breeding purposes, g ib b erellin alone did not produce sufficiently uniform bolting to be reliab le, and flow ering was not e a r lie r than following the conventional cold induction. Plants slow to respond to gibberellin in the greenhouse usually becam e very tall. T h eir flow ers, although appearing fairly norm al at anthesis w ere weak and produced little o r no seed. 7 to 10 Five treatm en ts of gibberellin at 100 ppm at day in terv als, sta rtin g when the plants had m axim um re s u lts . little flow ering. 6-8 tru e leaves, gave the F ew er treatm en ts resu lte d in some stem elongation but M ore trea tm e n ts re su lte d in ta lle r plants. Five applications of 1, 000 ppm re s u lte d in la rg e r and bushier plants, and flow ering was a w eek e a r lie r . L arge m atu re roots at tim e of firs t trea tm e n t responded v e ry ununiform ly in c o n tra st to plants f ir s t tre a te d at the 6-8 leaf stage. It is postulated that the annual habit is caused by the action of sev eral genes acting m ainly in a dominant fashion. However, the environm ent has a stro n g effect on th e ir expression and to evaluate the genotype of d iffer­ ent p lan ts would re q u ire controlled environm ents. It is apparent that by vigorous selection the genes for annual growth can be elim inated and non­ bolting stra in s obtained. LITERATURE CITED Abegg, F. A. 1936. A genetic factor fo r the annual habit in beets and linkage relatio n sh ip s. Jour. Agr. R es. 53: 493-511. ____________ j and F. V. Owen. 1936. A genetic factor fo r the annual habit in beets linkage relation ships. A m er. Nat. 70: 36. Blaney, L. T. 1957. G ibberellic acid shortens biennials flow ering tim e. O re. O rnam ental and N u rsery Digest 1: 5. Borthwick, H. A. et al. 1931. F lo ra l development in Daucus caro ta. A m er. Jour. Bot. 18: 784-806. _________________ , and S. L. E m sw eller. 1933. C arro t breeding e x p eri­ m ents. Proc. A m er. Soc. Hort. Sci. 30: 531-533. Boswell, V. R. 1929. Studies of p rem a tu re flower form ation in w intered over cabbage. Md. A gr. Exp. Sta. Bui. 313: 69-145. Bukovac, M. J . , and S. H. W ittwer. 1957. G ibberellin and higher plants. II. Induction of flow ering in biennials. Mich. A gr. Exp. Sta. Quart. Bui. 39: 650-660. C arolus, R. L. 1933. F lo ra l prim o r dia differentiation in beet and turnip. Proc. A m er. Soc. H ort. Sci. 30: 507-509. Chesnokov, V. A. 1934. The production of seeds of biennial plants by the m ethod of vernalization. Trudy L eningr. Obschch. E stesti. Otd. Bot. 65: 101-102. (English Summary). Chroboczek, E. A. 1934. A study of some ecological facto rs influencing seed sta lk developm ent in beets. Cornell Agr. Exp. Sta. Mem. 154. C lark, A. E. 1935. Inheritance of annual habit and mode of pollination in annual white sw eet clover. Jour. A m er. Soc. Agr. 27: 492-496. Dickson, M. H. 1956. Some effects of cold tem p eratu re on seedstalk p ro ­ duction in c a rro ts . Mich. State Univ. M. S. T hesis. 67. E m sw eller, S. L. 1928. Pollination and fertilizatio n in celery. Proc. A m er. Soc. H ort. Sci. 25: 29. ______ __________ • 1934. P rem atu re seeding in inbred lines of celery. Proc. A m er. Soc. Hort. Sci. 31: 155-159. F is h e r, J. E. 1956. Studies on the photoperiodic and th erm al control of flow ering in c a rro ts . Plant Phys. 31: Suppl. XXXYI. G askill, J. O. 1958. A p re lim in a ry re p o rt on the use of gibberellic acid to hasten reproduction and development in sugar beet seedlings. A m er. Soc. Sugar Beet Tech. 9: 521-528. G assn er, G. 1918. Beetrage zur physiologischen C h arak teristic som m erund w in terannueller G ew aschse inbesondere d er G etriebpflanzen. Z tsch r. Bot. 10: 419-480. G regory, F. G ., and O. N. P urvis. 1948. tem p era tu re . N ature 161: 859-860. Honma, S. R eversal of vernalization by high 1957. P ersonal comm unication - unpublished. Ito, J. H. 1957. Studies on the bolting of onion n . On the tem p eratu re for flow er bud form ation. Jour. Hort. A ssTn Japan. 25: 343-346. Johnson, J. J. 1942. C ro ss fe rtility relationships of "Golden Annual" sweet clover with common species of M elilotus. Jour. A m er. Soc. Agron. 34: 259-262. Kumaki, Y. 1956. V ernalization experim ent on c a rro ts . Japan. 25: 163-166. (English Summary) Jour. Hort. A ss'n Lang, A. 1956a. Stem elongation in a ro se tte plant, induced by gibberellic acid. Die N aturw issenshaften 43: 257-258. . 1956b. Induction of flower form ation in biennial Hyoscyanus by trea tm e n t with gibberellin. Ibid. 43: 284-285. . 1956c. G ibberellin and flower form ation. Ibid. 43: 544. . 1957. The effect of gibberellin on flower form ation. Proc. Nat. Acad. Sci. 43: 709-717. 68. M iller, J. C. 1929. A study of some fac to rs affecting seedstalk develop­ m ent in cabbage. C ornell Univ. Agr. Exp. Sta. Bui. 488. M unerati, O. 1931. L ’E red ita della tendenza annualita nella comune b arb ielo ta coltivata. Z. Zuchtung. A 17: 84-89. ____________. 1942. 177T-214T. The duration of the beet cycle. Int. Rev. A gric. 33: Paw ar, S. S ., and H. C. Thompson. 1950. The effect of age and size of plant at tim e of exposure to low tem p eratu re on reproductive growth in celery . Proc. A m er. Soc. Hort. Sci. 55: 367-371. Peto, F . H. 1934. The cause of bolting in Swede turnips. R es. 11: 733-750. Canadian Jour. Sakr, S. E . , and H. C. Thompson. 1942. Effect of tem p eratu re and photop erio d on seed stalk development in c a rro ts . Proc. A m er. Soc. Hort. Sci. 41: 343-346. . 1942. Effect of tem p eratu re and photoperiod on seedstalk developm ents in c a rro ts . C ornell Univ. Ithaca. Ph. D. T hesis. . 1944. E ffect of tem p era tu re and photoperiod on seedstalk developm ent in tu rn ip s. Proc. A m er. Soc. Hort. Sci. 44: 473-478. 1952. E ffect of tem p eratu re and photoperiod on seedstalk developm ent in p a rsn ip s. Ibid. 60: 299-300. Purvis, O. N ., and F . G. G regory. a tu re s. N ature 155: 113-114. 1945. D evernalization by high te m p e r­ . 1952. Study on vernalization XII. The re v e rsib ility by high tem p era tu re of the vernalized condition in Petkus w inter ry e. Ann. Bot. 16: 1-21. Smith, H. A. 1927. Annual vs biennial growth habit and its inheritance in M elilotus alba. A m er. Jour. Bot. 14: 129-146. Snedecor, G. W. 1946. S tatistical Methods. Iowa State College P ress, Am es, Iowa. 69. S tarrin g , C. C. 1924. P rem atu re seeding in celery. Exp. Sta. Bui. 168: 1-16. Univ. Montana Agr. Stevenson, J. M ., and W. J. White. 1937. An annual sweet clover (M. alba) of the dwarf branching type. Sci. Agr. 18: 210-215. Sutton, E. P. 260. 1924. Inheritance of bolting in cabbage. Jour. Hered. 15: 257- Thompson, H. C. 1929. P rem ature seeding in celery. Exp. Sta. Bui. 480: 1-50. C ornell Univ. Agr. ________________, and O. Smith. 1938. Seedstalk and bulb development in onion. C ornell A gr. Exp. Sta. Bui. 708. Van Heel, J. P. D. 9: 217-236. 1927. Inheritance of bolting in sugar beet. Genetica. V litos, H. J . , and W. Meundt. 1955. Interaction between vernalization and photoperiod in spinach. Contrib. Boyce Thompson Inst. 18: 159166. W ittwer, S. H ., and Bukovac, M. J. 1957. G ibberellin a new chem ical fo r crop production. Mich. Agr. Exp. Sta. Quart. Bui. 39: 469-494. _________ econom ic crops. . 1958. The effect of gibberellin on Economic Bot. 12: 213-255.