Seed Scarification Techniques in Silver Leaf Lupine

A major problem affecting the populations of the endangered mission blue butterfly of the northern coast of California is the declining numbers of its host plant, the silver leaf lupine (Lupinus albifrons), due to competition with non-native species. An attempt to ensure the restoration of populations of L. albifrons through germination in greenhouses and subsequent transplantation was initiated in 1984 and has continued to date within the Golden Gate National Recreation Area (GGNRA). This work analyzes the germination rates, necessary expertise, and time efficiency of three different scarification techniques applied to rupture the seed coat and induce germination, in a greenhouse setting. The three scarification techniques applied were: 1) manual nicking with a razor blade near the attachment point, 2) an acid bath in concentrated sulfuric acid for 30 minutes, and 3) scarification by agitation between two pieces of sandpaper for 5 minutes. Sample sizes of 98 seeds were studied at three different greenhouses in GGNRA, California. The experiment was controlled for soil composition and sowing procedure. The data provided no conclusive evidence for determining the most successful treatment for increasing L. albifrons germination rates. There was no significant difference between the germination rates of the three scarification treatments including the control but there was a strong difference in the rates of imbibition. The success of the control provides an impetus for further studies regarding the need for scarification on lupine seeds. Among the three scarification treatments, manual nicking had the highest imbibition rate, had the lowest cost and was the most time efficient. The significance of these results will aid in the restoration of L. albifrons to its native habitat and encourage increased populations of the mission blue butterfly. Introduction The threat of extinction for native species of North America has been a pressing issue since the time of European colonization, with the unchecked dominance of invasive species from across Europe. It is only in the past few decades that the environmental advantage of restoring native ecosystems has been recognized and the large task of restoration been undertaken (Mackay 1995, 2001). One such threatened species, the mission blue butterfly (Plebejus icarioides missionensis), native to the coast of northern California, has been federally listed as endangered since 1976 (Elliott 2003, pers. comm.). The population decline of this tiny blue butterfly is principally attributed to the loss of its host plant, the shrub-like silver leaf lupine (Lupinus albifrons ssp. collinus) (La Pierre 1998). Figure 1. Mission blue butterfly with L. albifrons The mission blue butterfly (Figure 1) relies on one of three species of lupine to lay its eggs and for nourishment of larvae (Arnold, 1994). Due to a decline in numbers resulting from a loss of habitat, restoration of lupine to its native ecosystem is necessary in order to increase butterfly populations. In 1984 a recovery plan drawn up by the U.S. Fish and Wildlife Service (USFWS), outlined the need to protect mission blue habitat and to repair habitat damaged by urbanization, off highway vehicle traffic, and invasion by exotic, non-native plants (Arnold 1987). This recovery plan advocates the restoration of the native habitat of the mission blue butterfly by planting high numbers of native species of lupine, including L. albifrons. Due to development and construction, what is left of the native habitat of the mission blue butterfly is mainly the area that comprises the Golden Gate National Recreation Area (GGNRA). Within this protected area, the survival of the silver leaf lupine is impeded by non-native species and seed predation. This competitive environment has provided an impetus for the development of new methods to reestablish the lupine habitat. To decrease seed predation, biologists at GGNRA in the past, have collected lupine seeds in the field, germinated them in greenhouses and later transplanted them back to the field once they have grown to a substantial height (Setty 2003, pers. comm.). But low greenhouse germination yields have called for an inquiry into better methods for seed germination of the silver leaf lupine. Past studies have been done with seeds of other classes of Lupinus to increase germination rates (Kaye 2001), but there is no information on studies of L. albifrons. Of the techniques applied to bush lupine (L. sulphureus), which included a variety of scarification and temperature treatments, the method which produced the highest percentage of germination was nicking the seed coat with a razor blade before sowing (Kaye 2001). Scarification techniques, such as nicking or the use of sandpaper, are meant to simulate the scarification of the seed coat that would occur in the natural habitat of the coastal sand dunes (Setty 2002, pers. comm.). A comprehensive experiment that specifically studies different scarification methods and the correlating germination percentages for silver leaf lupine seeds will facilitate lupine restoration, and in turn, the restoration of mission blue butterfly populations. The objective of this research was to determine a scarification technique that would produce the highest yield of germination for the silver leaf lupine in the GGNRA. Three different scarification methods were tested: manual nicking, acid bath washing (Simms 2002) and sandpapering (Natoli 2003). While recording imbibition and germination rates, other factors were noted, including time efficiency of each treatment and necessary expertise. Based upon past studies of different classes of lupine (Sholars 2002), my hypothesis was that nicking seeds near the attachment point would produce the highest percentage of seed germination, but would also be the most labor intensive. To determine the scarification technique that is best for the entire GGNRA, treated seeds were sown in three different greenhouses throughout the Recreation Area. This study thus yields results for germination that can be compared across greenhouses, to determine the method that is best overall, or that is particularly best for a specific greenhouse. This technique can then be the basis for more efficient restoration projects of the silver leaf lupine at all greenhouses in GGNRA. Methods All L. albifrons (LUAL) seeds used in this experiment were purchased by the GGNRA from S & S Seeds, collected from Los Alamos, California. The effects of scarification were examined with three different scarification techniques: Manual nicking The nicking of seeds was performed using a one sided razor blade. The seeds were nicked on the side of the radicle, near the attachment point, and a small portion of the seed coat was removed. The dried weight of the seeds was taken, after which the seeds were soaked in tap water for one hour, dried and then weighed to confirm that they had imbibed water, signified by a 50% increase in weight (also signified by a transparent seed coat). After imbibition was confirmed, seeds were sown (Setty and Heimbinder 2002, pers. comm.). Acid bath Seeds in batches of 25 were counted into small stopped glass flasks of 25 ml. In a chemical hood, concentrated sulfuric acid was poured onto seeds in sufficient volume to cover seeds with some excess. The flasks were swirled occasionally over a period of 25-35 minutes. The acid was then poured off into a glass beaker and water was immediately added to the seeds in the flask, which was then capped and shaken. Water was then poured off into a second beaker, more water was added, and the flask was again capped and shaken; repeating the rinsing in this manner five more times with each batch. Seeds were then soaked in water and refrigerated overnight at 4 ̊ C and then sown the next day (Simms 2002, pers. comm.). Sandpaper Seeds in groups of 25 were placed on the bottom of a metal sieve which was lined with sandpaper (60 coarse grit). Using a wooden block covered in sandpaper of the same grade, the seeds were rubbed in a circular motion for 5 minutes between the two pieces of sandpaper. The seeds were then soaked for one hour, dried, and sown. The experiments were performed at three greenhouses in GGNRA: (1) Fort Funston, San Francisco, (2) Presidio, San Francisco, and (3) Headlands, Marin. The soil standard at all greenhouses was a 3:1 mixture of GGNRA compost mix to perlite for all treatments; in addition, one set of nicked seeds was potted in 100% Sunshine Mix #4 at the Headlands nursery. This variable in soil composition was tested upon the request of the GGNRA native plant nursery staff since the mix is easily available and its success with lupine seeds is unknown. Due to space and monetary limitations at the Fort Funston and Presidio greenhouses, only the sandpaper and nicked treatment were tested. At the Marin Headlands greenhouse, which is the newest and largest facility, all treatments were tested including a control with no scarification applied. In total, three different treatments plus a control, with no treatment applied, were tested, and one soil variation. At each location, a sample size of 98 seeds was used for each treatment. All plants were sown in Stuewe & Sons brand Ray Leach “Stubby” Cells (1.5”x5.5”), one seed per pot, in racks with a capacity of 98 (14 x 7). All pots were sterilized before use by being dipped in Clorox water (10% solutions for 30 seconds) before sowing and then rinsed with water. Seeds were sown at a depth of 0.5cm. At Fort Funston and the Presidio, the pots were checker boarded. At Headlands the pots were staggered by placing them in the rack in an ordered sequence (treatments 1-5) up and down each column. Plants at Fort Funston and Headlands greenhouses were protected from predation by being placed inside of wire mesh enclosure cages. At the Presidio, where predation is not a problem, a plastic mesh sheet was placed over the racks. Germination was determined by the appearance of true leaves (Young 2002). All seeds at the Fort Funston and Presidio greenhouses were placed under an automatic irrigation system which was set according to the discretion of the particular nursery manager, keeping soil barely moist, but never dry. Seeds at the Headlands greenhouse were watered with a misting system which keeps soil constantly damp at a setting of 25 VPDs (vapor pressure deficit). The temperature at each greenhouse was monitored with a digital temperature recorder which was placed inside the cages for a three week period. Each recorder was hung inside an upsidedown plastic cup to protect if from water damage due to constant watering. These recorders produced temperature graphs which facilitate the comparison of the temperature differences in the three greenhouses. Germination data was collected and compiled into tabular form, documenting the germination progress for all treatments. Results Treatment Requirements and Imbibition Rates The manual nicking treatment for 98 seeds requires approximately 40 minutes; the average person can nick approximately 150 seeds per hour with little prior training. The only cost required for this treatment is the initial purchase of the razor blades, which is relatively minimal. Nicking yielded an 80% imbibition rate after soaking for one hour. For the acid scarification of a total number of 98 seeds, three hours of labor were required. The acid treatment must be performed under a fume hood using concentrated sulfuric acid, an extremely dangerous substance and necessitates training in laboratory safety. Sulfuric acid must be bought in a large quantity and is relatively expensive. This treatment yielded a 63% imbibition rate. The sandpaper treatment required 5 minutes to scarify 98 seeds and can be performed with no prior experience. A large quantity of sandpaper can be purchased at a very low cost. This technique resulted in a high percentage of destroyed seeds, yielding a 32% imbibition rate. Imbibition indicates that the seed coat has been successfully scarified which is necessary for germination of seedlings (Young 2002). Only seeds that displayed visible signs of imbibition, determined through the display of a transparent seed coat and increased size, were sown. Figure 2 demonstrates that there was a significant difference (p < 0.000) in the imbibition rates between the three treatments. The control had a 1% imbibition rate, but was not included in the figure.