Lin, W.C. 1984. The effect of soil cooling and derzoek in Nederland, Aalsmeer, The Wilkins, H. F., W.E. Healy, and T.L. Gilbertson- high intensity lighting on flowering of Alstroe- Netherlands. Ferriss. 1980. Comparing and contrasting the control of flowering in Alstroemeria ‘Regina’, meria ‘Regina’. HortScience 19(4):515-516. Vonk Noordegraaf, C. 1975b. Temperature and Powell, M.C. and A.C. Bunt. 1986. The effects Freesia × hybrids and Lilium longiflorum, p. daylength requirements of Alstroemeria. Acts 51-63. In: C. Brickell, D.F. Cutter, and M. of propagation date on production in Al- Hort. 51:267-274. stroemeria ‘Campfire’ and ‘Red Sunset’. Scien- Gregord (eds.). Petaloid monocotyledous, hor- tia Hort. 28:147–157. Vonk Noordegraaf, C. 1981. Bloemproduktie bij ticultural and botanical research. Academic, New Vonk Noordegraaf, C. 1975a. Alstroemeria: Be- Alstroemeria ‘Walter Fleming’. PhD Diss., Agr. York. lichting bij ‘Orchid’, p. 15. In: Bloemisterij on- Univ., Wageningen, The Netherlands.

HORTSCIENCE 25(2):191-193. 1990. and percentage were investigated. Eight seed lots were obtained from pro- ducers during Sept. -Nov. 1985 (Table 1) and Temperature Requirements for Seed stored in paper containers in the laboratory ( » 22C). Seeds had been held in warehouse Germination of Three storage by producers. The experiments de- scribed were conducted between Jan. and June Species’ 1986. Viability for each seed lot was determined Phil S. Allen1 and Susan E. Meyer using tetrazolium staining (Grabe, 1970; Division of Wildlife Resources, USDA Forest Service Intermountain Kitchen, 1988). Two 50-seed replications were imbibed on moist blotters for 12 hr, Research Station, Shrub Sciences Laboratory, 735 N. 500 East, Provo, pierced with a needle, and immersed in a 1% UT 84606 (w/v) solution of buffered tetrazolium chlo- Additional index words. Penstemon eatonii, Penstemon palmeri, Penstemon strictus, ride. After 48 hr, seeds were bisected lon- gitudinally and examined under a dissecting perennials, seed dormancy, stratification, wildflowers microscope. Seeds with embryos stained Abstract. To determine optimum germination temperatures and effective dormancy- completely red were classed as viable; re- breaking procedures, field-grown (1983-85) seeds of ‘Bandera’ Rocky Mountain pen- maining seeds were classed as nonviable. stemon (Penstemon strictus Benth), ‘Cedar’ Palmer penstemon (Penstemon palmeri Gray), Two germination experiments were con- and firecracker penstemon (Penstemon eatonii Gray) were subjected to various cold ducted. Treatments consisted of four repli- stratification and incubation temperature treatments. Increased germination following cations of 50 seeds each, incubated on two an 8-week stratification occurred in seed lots containing dormant seeds, but a 2-week stratification generally failed to break dormancy. Older (1983) seeds of ‘Bandera’ and ‘Cedar’ penstemon germinated to full viability without stratification. All species showed a marked decrease in germination percentage above 20C; 15C consistently produced maximum germination after 4 weeks. At 15C, mean times to 90% of total germination were 11, 22, and 29 days for ‘Bandera’, ‘Cedar’, and firecracker penstemon, respec- tively. Transfer of seeds failing to germinate at warm temperatures (25 and 30C) to

15C and applying 720 µM gibberellic acid (GA3) solution was effective in breaking primary dormancy of firecracker penstemon and secondary dormancy of ‘Bandera’ penstemon. Our findings suggest that incubation below 20C, combined with 8 weeks of stratification or the use of after-ripened seed, may improve seed propagation efforts for these species.

Widespread interest in perennials and in effects on viability and dormancy, and ef- wildflowers for use as low-maintenance or- fective dormancy-breaking procedures would namental has increased marketing oppor- benefit seed propagation efforts. tunities for native (Aimone, 1986; Penstemon spp. offer considerable poten- Allen, 1985; Cox and Klett, 1984; Gilbert, tial for increased horticultural use. There are 1987; Otteson, 1986). Unfortunately, seed more than 250 perennial species of native dormancy resulting in poor or sporadic ger- North American (Cronquist et mination can discourage growers from sup- al., 1984), a few of which are currently cul- plying desirable materials to the public tivated (Kelaidis, 1986). While seed dor- (Aimone, 1986). For such species, infor- mancy occurs in this genus, few reports on mation on germination requirements, seed age germination have been published (Atwater, 1980; Cox and Klett, 1984; Maguire and Overland, 1959; Salac and Hesse, 1975; Salac and Traeger, 1982). Received for publication 7 Mar. 1989. This re- In this study, we compared laboratory ger- search was supported in part by Pittman–Robert- mination characteristics for successive yearly son Federal Aid to Wildlife Project W82-R and harvests of three penstemon species recently by funds provided by the Utah Dept. of Agricul- brought into field cultivation for seed pro- ture. The cost of publishing this paper was de- duction: ‘Bandera’ Rocky Mountain penste- frayed in part by the payment of page charges. mon (Hooks and Oaks, 1982), ‘Cedar’ Palmer Under postal regulations, this paper therefore must be hereby marked advertisement solely to indicate penstemon (Stevens and Monsen, 1988), and Fig. 1. Effect of temperature on seed germination this fact. firecracker penstemon. In particular, the ef- of (A) ‘Bandera’, (B) ‘Cedar’, and (C) fire- lCurrent address: Dept. of Horticultural Science, fects of cold stratification and of incubation cracker penstemon. Curves represent fitted po- Univ. of Minnesota, St. Paul, MN 55108. temperature on laboratory germination rate lynomial equations (P < 0.001) for each seedlot.

HORTSCIENCE, VOL. 25(2), FEBRUARY 1990 191 Table 1. Source data for experimental seedlots ofPenstemon. Most firecracker penstemon seeds were Common and Latin Location Production Seed class and dormant for both harvest years (Table 2). names grown years source While 8 weeks of stratification broke dor- mancy in the 1984 seed lot, nearly half of ‘Bandera’ Rocky Meeker, Colo. 1983-85 Foundation seed; Mountain penstemon, Upper the 1985 seeds were still dormant following Penstemon strictus Environmental Plant this treatment. Materials Center All species and all seed lots germinated (SCS)z, Meeker, Colo. maximally after 4 weeks at 15C (Fig. 1). In ‘Cedar’ Palmer Ephraim, Utah 1983-85 Foundation seed; general, germination was drastically reduced penstemon, Great Basin at higher temperatures. Reduced 4-week ger- Penstemon palmeri Experimental Station mination at 5 and 10C is due to a slower (USFS)y, Ephraim, Utah germination rate at low temperatures. After Firecracker Caldwell, 1984-85 Commercial seedlots; 8 weeks, germination at 5 and 10C was penstemon, Native Plants, Inc., equivalent to that in the 15C treatment (data Penstemon eatonii Salt Lake City, Utah not shown). zSoil Conservation Service. Of the three species, ‘Bandera’ had the yU. S. Forest Service. highest relative germination at all tempera- tures (Fig. 1A). Temperature sensitivity, as indicated by reduced germination at temper- Table 2. Absolute percent germinationz of Penstemon seeds after stratification (5C) for 0, 2, or 8 atures above and below the optimum, ap- weeks followed by incubation (15C) for 10, 8, or 2 weeks, respectively. pears to be related to the degree of seed dormancy. The nondormant 1983 seed lot Type of Viability Stratification (weeks) penstemon Year (%)y 0 2 8 germinated well over a much wider range of temperatures than did the 1984 and 1985 seed Germination (%) lots, which contained dormant seeds. Bandera 1983 95 92 93 90 1984 93 79* 93 98 Germination of ‘Cedar’ was highly tem- 1985 94 50** 68** 82* perature-sensitive and was completely pre- Cedar 1983 78 76 54* * 83 vented at 30C (Fig. 1B). Relative germination 1984 88 68** 74* 84 of the 1984 seed lot (containing dormant 1985 63 56 58 56 seeds) was generally lower than for the other Firecracker 1984 76 24** 15** 65 two lots, particularly at 10 and 25C. This 1985 87 24** 34* * 46** result suggests that temperature sensitivity zValues within rows significantly different from viability control at P = 0.05 (*) and 0.01 (**). increases with increased seed dormancy in yViability determined by tetrazolium staining; see text for details. this species as well. Firecracker penstemon seeds did not ger- minate above 15C (Fig. 1C). Seeds har- vested in 1984 germinated more rapidly at layers of germination blotting paper in cov- razolium viability included as a measure of lower temperatures than did 1985 seeds (data ered petri dishes and moistened as needed. maximum possible germination for compar- not shown), although neither seedlot had Dishes were placed in random arrangement ative purposes. For the second experiment, germination percentages that approached tet- in dark germinators. Germinated seeds (rad- germination data were converted to a relative razolium viability estimates. icle emerged to 1 mm) were counted and (percent of viable seed) basis to facilitate Poor germination at 25 and 30C prompted removed under cool-white fluorescent lights comparisons among seed lots. Temperature efforts to determine whether ungerminated at least weekly. Light was probably neither response curves were determined for each seeds remained viable following exposure to strongly limiting nor inhibiting under these seed lot using polynomial regression tech- supraoptimal temperatures for 4 weeks (as conditions (S. E. M., unpublished data). niques. Time to 90% of total germination indicated by germination following transfer

In the first experiment, seeds were sub- (T90) at the optimum temperature (i.e., tem- to 15C and moistening with either water or jected to three cold-stratification treatments: perature resulting in maximum 4-week ger- GA3 solution). For ‘Bandera’, a 4-week in- 15C for 10 weeks (no stratification), 5C for mination) was determined for each seed lot cubation at supraoptimal temperatures re- 2 weeks followed by 15C for 8 weeks (2- by interpolating on plotted rate curves. sulted in a 20% decrease in total germination week stratification), and 5C for 8 weeks fol- ‘Bandera’ harvested in 1983 germinated to following transfer to 15C and moistening with lowed by 15C for 2 weeks (8-week stratifi- full viability without stratification (Table 2). water only (Table 3). However, when moist- cation). Seed harvested in 1984 and 1985 responded ened with the GA3 solution, germination In the second experiment, seeds were in- favorably to stratification treatments, al- percentage was not significantly different from cubated at a constant 5, 10, 15, 20, 25, or though the 1985 seedlots failed to germinate the control. This change suggests that ex- 30C for 4 weeks. At the end of 4 weeks, to full viability even with an 8-week strati- posure to high temperatures during imbibi- ungerminated seeds from the 25C and 30C fication. tion can induce secondary dormancy, i.e., treatments (warm temperatures) were trans- Stratification enhanced germination of “thermodormancy” (Bewley and Black, ferred to 15C on blotters moistened either ‘Cedar’ only for 1984 seeds (Table 2). Seeds 1982, p. 328), which can be overcome by with water of 720 µM gibberellic acid (GA3) harvested in 1985 had low viability, were not application of GA3 in this species. solution. Ungerminated seeds in the 15C dormant, and showed no response to strati- In contrast to ‘Bandera’, seeds of ‘Cedar’ treatments were left in the initial test con- fication. Nondormant 1983 seeds showed and firecracker penstemon exposed to high ditions to serve as controls. depressed germination following a 2-week temperatures showed no reduction in total All germination and viability data were stratification, but germinated to full viability germination following removal to 15C. Ap- arcsin-transformed before analysis of vari- with the 8-week stratification treatment. Re- parently, failure of these seeds to germinate ance to decrease variance heterogeneity as- duced germination following exposure to low at high temperatures was due only to en- sociated with binomial data (Little and Hills, temperature has been described as a type of forced, or relative, dormancy (Bewley and 1978, p. 159); original data are reported. secondary dormancy (Baskin and Baskin, Black, 1982, p. 63). Of interest is the marked Dunnet’s test for comparing treatment means 1985; Bewley and Black, 1982, p. 308). increase in germination percentage of fire- with controls was used where appropriate. Kitchen (1988) has reported this phenome- cracker penstemon following application of

Absolute germination percentages were used non in other seedlots of this species and in GA3 (60%. as compared with 22% in the con- for analysis of the first experiment, with tet- several other penstemons. trol). Evidently, primary dormancy of fire-

192 HORTSCIENCE, VOL. 25(2), FEBRUARY 1990 Table 3. Relative percent germination of Penstemon seeds incubated at 15, 25, or 30C for 4 weeks, tinuum. Bioscience 35(8):492-498. followed by incubation at 15C for 1 week. During the 5th week, seeds were moistened with water Bewley, J.D. and M. Black. 1982. Physiology

or 720 µM GA3 solution. and biochemistry of seeds in relation to ger- mination: 11. Viability, dormancy, and environ- Moistening agent during 5th week mental control. Springer-Verlag, Berlin. Water only 720 µM GA3 Cox, R.A. and J.E. Klett. 1984. Seed germination Four-week incubation temperature (C) requirements of native Colorado plants for use in the landscape. Plant Prop. 30(2):6-10. Type of 15 25 30 25 30 z y Cronquist, A., A.H. Holmgren, N.H. Holmgren, penstemon Germination (%) J.L. Reveal, and P.K. Holmgren. 1984. lnter- Bandera 88 68* 67* 87 90 mountain flora. vol. 4. New York Botanical Cedar 63 67 65 71 65 Garden, Bronx. Firecracker 22 24 --- 60** --- Gilbert, B. 1987. In from the fields, wildflowers zFor each type of penstemon, values are averaged across years of collection. find anew welcome among gardeners. Smith- *,**Values within rows significantly different from 15C control at P = 0.05 and 0.01, respectively. sonian 18(1):37-45. Grabe, D.F. (cd.). 1970. Tetrazolium testing handbook for agricultural seeds. Handbook on cracker penstemon seeds was largely broken means of obtaining more uniform germi- Seed Testing. Contrib. 29. Assn. Offic. Seed Anal. by return to 15C and moistening with GA3 nation. solution. An alternative to stratification for ‘Ban- Hooks, R.F. and W.R. Oaks. 1982. ‘Bandera’ As determined by interpolation from plot- dera’ and ‘Cedar’ penstemons is to use “after- Rocky Mountain penstemon. HortScience 17:683. ted rate curves (not shown), mean T90 values ripened” seed (seed that has lost dormancy at 15C were 11, 22, and 29 days for ‘Ban- during dry storage). For both of these pen- Kelaidis, P. 1986. Penstemons. Pacific Hort. dera’, ‘Cedar’, and firecracker penstemon, stemons, seed harvested in 1983 (» 2½ years 48(2):28-34. respectively. Poor germination (low per- before being used in this study) had lost all Kitchen, S.G. 1988. Germination studies with fif- teen species of Intermountain penstemons. MS centage and slow rate) of firecracker penste- dormancy. Stevens et al. (1981) have re- Thesis, Brigham Young Univ., Provo, Utah. mon increases the difficulty of seed ported high viability for both of these species Little, T.M. and F.J. Hills. 1983. Agricultural ex- propagation. For this species; use of 1-year- following several years of warehouse stor- perimentation: Design and analysis. Wiley, New old seed did not preclude the need for strat- age. York. ification, and, even with stratification, ger- Based on the above results, ‘Bandera’ and Maguire, J.D. and A. Overland. 1959. Laboratory mination remained sporadic. Treatment with ‘Cedar’ appear relatively easy to germinate germination of seeds of woody and native plants. GA3 solution may be an effective means of as long as suitable temperatures are pro- Wash. Agr. Expt. Sta. Circ. 349. breaking dormancy in this species. How- vided. Firecracker penstemon presents more Otteson, C. 1986. Landscape perennials are trans- ever, further research to develop suitable difficulty because of primary dormancy that formed by changing seasons. Amer. Nursery- techniques is needed, as we have noted that is not substantially reduced during storage. man 163(8):72–75. seedlings resulting from GA3-treated seeds Results of experiments with wild-collected Plummer, A. P., D.R. Christensen, and S.B. were spindly and difficult to maintain under seed (Kitchen, 1988) suggest that it maybe Monsen. 1968. Restoring big game range in greenhouse conditions. possible to select a firecracker penstemon Utah. Utah State Dept. Fish and Game Publ. In the field, penstemons are typically cultivar with lower seed dormancy, thereby 68-3. seeded in the fall (Hooks and Oaks, 1982; improving seed propagation. Salac, S.S. and M.C. Hesse. 1975. Effects of Plummer et al., 1968; Salac and Traeger, storage and germination conditions on the ger- 1982: Stevens and Monsen, 1988). Pre- mination of four species of wildflowers. J. Amer. Literature Cited Soc. Hort. Sci. 100:359-361. sumably, the winter environment naturally Aimone, T. 1986. Success with perennials. Grower- provides cold stratification. Artificial strat- Salac, S.S.and J. M. Traeger. 1982. Seeding dates Talks 49(11):125-133. and field establishment of wildflowers. Hort- ification may be necessary for seed prop- Allen, P.S. 1985. Studies on seed performance of Science 17:805-806. agation under greenhouse conditions. In this five arid landscape species. MS Thesis, Brigham Stevens, R., K.R. Jorgensen, and J.N. Davis. 1981. study, no germination occurred during 2 Young Univ., Provo, Utah. Viability of seed from 32 shrub and forb species weeks of stratification. However, between Atwater, B.R. 1980. Germination, dormancy, and through 15 years of warehouse storage. Great 50% and 100% germination in the 8-week morphology of seeds of herbaceous ornamental Basin Naturalist 41:274-278. stratification treatments occurred during plants. Seed Sci. & Technol. 8:523-573. Stevens, R. and S.B. Monsen. 1988. ‘Cedar’ Pal- stratification. Thus, stratification treat- Baskin, J. M. and C. C. Baskin. 1985. The annual mer penstemon: A selected penstemon for semi- ments are not perceived as an effective dormancy cycle in buried weed seeds: A con- arid ranges. Rangelands 10:163–164.

HORTSCIENCE, VOL. 25(2), FEBRUARY 1990 193