PERCUSSION AS AN ALTERNATIVE SEED TREATMENT FOR
ROBINIA NEOMEXICANA (NEW MEXICO LOCUST)
BY
NABIL Y. KHADDURI
A thesis submitted to the Graduate School
in partial fulfillment ofthe requirements
for the degree
Master of Science
Major subject: Horticulture
New Mexico State University
Las Cruces, New Mexico
May 2002 "Percussion as an Alternative Seed Treatment for Robinia neomexicana (New Mexico
Locust)," a thesis by Nabil Y. Khadduri, in partial fulfillment ofthe requirements for the degree, Master of Science, has been approved and accepted by the following:
Interim Dean of the Ora uate School
Chair ofExamining Committee
, Date
Committee in charge:
Dr. John T. Harrington, Chair
Dr. David R. Dreesen
Dr. Leigh Murray
Dr. Rolston St. Hilaire ACKNOWLEDGMENTS
I would like to thank Dr. John Harrington for his guidance, sense ofhumor and flexibility throughout my studies. I would also like to thank the members ofmy committee for their support, especially Dr. Murray, who patiently taught me the subtleties of categorical analysis of variance.
I am also grateful to Dr. Anne Wagner and Molycorp Mine, who provided funding for this research. Thanks to Dr. Bob Karrfali for his assistance with x-ray radiography and to Mohamed Murshed and Dr. Soumitra Ghoshroy for their assistance with scanning electron microscopy. Special thanks go to my colleague Lee
Rosner, who assisted me in all phases ofthis project. Finally, I thank my family for their enduring love: Jill, Walid, Amin, Rolla, Khalid, Piet, Zach, Charlie and above
all, my'wife, Lucie.
111 VITA
June 3, 1972 Born in Beirut, Lebanon.
June 1990 Graduated from Georgetown Day High School, Washington, DC.
May 1994 Graduated from Pomona College, Claremont, CA with Bachelor ofArts Degree in Political Science.
July 1995 Graduated from University ofNew Mexico, Albuquerque, NM with Post-baccalaureate Teaching Degree.
August 1994- March 1998 6th Grade Teacher,Banta Fe Public Schools, Santa Fe, NM and Environmental Educator, Open Space Division, Albuquerque, NM.
March 1998-August 2000 Assistant Manager, Santa Ana Native Plant Nursery, Santa Ana Pueblo, NM.
August 2000-May 2002 Graduate Research Assistant at New Mexico State University
FIELD OF STUDY
Horticulture
iv ABSTRACT
PERCUSSION AS AN ALTERNATIVE SEED TREATMENT FOR
ROBINIA NEOMEXICANA (NEW MEXICO LOCUST)
BY
NABIL Y. KHADDURI
Master of Science in Horticulture
New Mexico State University, 2002
Las Cruces, New Mexico
Dr. John T. Harrington, Chair
New Mexico locust (Robinia neomexicana A. Gray), a small tree native to montane regions of the southwestern United States, fills a successional role in post disturbance situations. The ability to colonize rapidly and ameliorate harsh sites contributes to the importance ofNew Mexico locust in reclamation efforts. Seedling production for outplanting on disturbed sites has been hampered by poor germination.
This research sought to improve seed propagation techniques in New Mexico locust.
Nine seed sources were collected across a range of latitudes within New
Mexico. Seed treatments included a control, three levels of hot water and four levels of percussion scarification. Hot water treatments included immersion at 100°C, and boiling for periods of 30 seconds and 60 seconds. Percussion treatment levels were 1, 2 , 4 , and 8 minutes. A total offour lOO-seed samples were used to test each source
by scarification treatment combination. Categorical Analysis of Variance was used to
determine main effects and interactions. Pairwise comparisons were conducted to
compare non-control treatment means using a conservative alpha value of 0.05
divided by the number of comparisons.
Averaged over all sources, and for each source, percussion scarification
durations of4 or 8 minutes yielded greatest germination. In addition to consistently
bigh germination, 4- and 8-minute percussion-treated seeds germinated as fast or
faster than all other treatments. Averaged over all sources, germination increased
with increasing durations ofpercussion, then leveled off for the 4- and 8-minute
treatments. Immersion and 30-second boiling scarification maximized germination
for the hot water treatments, with a decrease in germination when seeds were boiled
for 60 seconds.
Percussion scarification specifically weakens the strophiole, the natural
dormancy barrier in many papilionoid legumes, without excessively damaging the
seed coat. As with New Mexico locust, papilionoid legumes often play an integral
role in the revegetation ofdisturbed lands. The percussion scarification technique
developed in this research has the potential to facilitate the use ofthese valuable
species in reclamation projects.
VI CONTENTS
LIST OF TABLES...... ix
LIST OF FIGURES ...... ,...... · .. ···.. ··········.. ··..... x
INTRODUCTION...... 1
LITERATlTRE REVIEW ...... '" ...... , .. , ...... 4
Seed Donnancy...... 4
Physical Donnancy...... 5
Evolutionary Role ofPhysical Donnancy...... 6
Methods for Breaking Donnancy...... 6
Acid Scarification...... 6
Hot Water Scarification...... 8
Mechanical Scarification ...... , ...... 11
Alternative Method for Breaking Donnancy: Percussion...... 12
Objectives...... 15
Hypothesis...... :...... '"...... , ...... 15
MATERIALS AND METHODS...... 16
Seed Collection and Cleaining...... 16
Experimental Design and Treatment Structure...... 16
Treatments...... ~ •• ,. •••• * ...... - ..... * •• ~ ...... "'~ ...... 20
Gennlnatlon . . T estmg ...... 21
X-ray Radiography ...... 22
Scanning Electron Microscope Photography ...... ,...... 22 Analysis...... 22
RESULTS...... 24
DISCUSSION...... 36
Detennining the Optimal Treatment...... 36
Strophiole and Seed Coat Behavior...... 37
CONCLUSION...... 44
Appendices
A. PILOT STUDY: PERCUSSION AS AN ALTERNATIVE SCARIFICATION FOR NEW MEXICO LOCUST AND BLACK LOCUST SEEDS...... 47
B. GERMINATION PERCENTS AND STANDARD ERRORS FOR SOURCE, BLOCK AND TREATMENT FOR NEW MEXICO LOCUST...... 64
LITERATURE CITED...... 72
Vlll LIST OF TABLES
Table Page.
1. Latitude, longitude, location, elevation and collection date ofNew Mexico locust seed sources...... 18
2. Mean seed weight and standard error by source...... 19
3. Categorical Analysis ofVariance table for, germination response to seed treatment, seed source, block and the interaction ofthe factors...... 25
4. Effect of seed source and treatment on G50 (days to 50% germination)...... 32
B1. Germination percents and standard errors for source, block and treatment for New Mexico locust...... 65 LIST OF FIGURES
Figures Page
1. Seedsourcecollectionmap...... 17
2. Effect ofNew Mexico locust seed source on germination for data averaged over scarification treatments...... 26
3. Effect of scarification on germination for New Mexico locust for data averaged over all seed sources...... 27
4. Effect of seed source by treatment interaction on New Mexico locust germination for percussion scarification, pattern a and pattern b...... 28
5. Effect of source by treatment interaction on New Mexico locust germination for hot water scarification...... 29
6. Timecourse of New Mexico locust germination following scarification treatments...... 31
7. X-ray radiographs of Sacramento seed source perc ussed for 1, 4 or 8 minutes...... 33
8. Seed damage close-ups...... 34
9. SEM images of untreated vs. 4-minute percussion-treated strophiolar cells ofNew Mexico locust seeds...... 35
10. Strophiole ofunimbibed New Mexico locust seed (a) and following water uptake (b, c)...... 38
11. Damage to New Mexico locust seeds observed during germination counts following hot water scarification...... 40
12. Effect of hot water scarification (a) and percussion scarification (b) on interaction of Jemez and Raton seed sources...... 42
New Mexico locust: a primary invader following disturbance...... 50
Implementation of hot water and percussion scarification...... 52
Effect of scarification on germination response for New Mexico locust ...... '" ...... 55 x A4. Effect of selected scarification treatments on germination speed for New Mexico locust...... 56
AS. Effect of scarification on germination response for black locust...... 57
A6. Effect of selected scarification treatments on germination speed for black locust...... 59
A7. Damage to New Mexico locust seeds observed during germination counts following hot water scarification...... 60
AS. New Mexico locust seeds following percussion treatment at time of germination counts...... 61
A9. Strophiole ofunimbibed New Mexico locust seed (a) and following water uptake (b,c)...... 63 INTRODUCTION
New Mexico locust (Robinia neomexicana A. Gray), a small tree native to the
southwestern United States, occurs at elevations from 1,200 to 2,800 meters. This
species fills a successional role in post-disturbance situations. As a primary invader,
New Mexico locust quickly establishes on burned areas arid flood banks, as well as
road cuts (Wagner et aI. 1992). Rapid growth, crown sprouting and prolific root
suckering favor the successful establishment ofNew Mexico locust on disturbed sites
(Simpson 1988, Gottfried 1980). These attributes also may make New Mexico locust
well suited for steep-slope revegetation where erosion is a problem.
New Mexico locust is a nitrogen-fixing legume that tolerates and improves
nutrient-poor soils. Stands ofNew Mexico locust increase Nitrogen (N) and Calcium
(Ca) in the forest floor. Levels of Carbon (C), Phosphorous (P), Sulfur (8) and
Potassium (K) aIso have been shown to increase in soil beneath New Mexico locust
(Klemmedson 1994).
The ability to colonize rapidly and ameliorate harsh sites contributes to New
Mexico locust's potential as a nurse plant. A nurse plant colonizes an inhospitable
site and creates an environment suitable for successional plant establishment. New
Mexico locust demonstrates this role in the ponderosa pine (Pinus ponderosa Douglas
ex Lawson & C. Lawson)/Gambel oak (Quercus gambelii Nutt.) community. After a
disturbance such as wildfire, New Mexico locust and Gambel oak colonize and
dominate the site until shaded out by ponderosa pine. Once shaded, these species become understory shrubs, a process taking an average of 15-20 years (Dick-Peddie 1993, Hanks and Dick-Peddie 1974). New Mexico locust gradually declines, with small dense patches averaging about 0.05 ha beneath pine, until the next disturbance once again offers it a competitive advantage (Klemmedson 1994).
Gottfried (1980), expecting ponderosa pine regeneration to be greater where
New Mexico locust had been eradicated, found that survival of planted pine seedlings was greater where locust had not been removed. In a follow-up study, Gottfried
(1980) noted that soil moisture in the top 57 cm was highest in 5-year old locust sites, as compared to grass or 20-year old locust sites. He concluded that managing an appropriate cover ofNew Mexico locust could help regenerate pine, the later successional species.
The ability ofNew Mexico locust to improve harsh sites also makes it a candidate for reclaiming disturbances associated with mining. Black locust (Robinia pseudoacacia 1.) has been used for decades to reclaim mine spoils and other disturbed sites in the eastern US and around the world (Keresztesi 1988, Zimmerman and Carpenter 1980). Ashby et al. (1985) described similar positive attributes of black locust also mentioned previously for New Mexico locust: quick cover for stabilization, supply ofN and nutrient-rich litter to the soil, and site improvement for establishment oflater successional trees. Referring specifically to mine reclamation, the authors note the ability of black locust to grow on a wide range ofmine soil conditions, including extremely acid soils. Black locust also shows some tolerance to soils compacted by grading and topsoiling practices (Ashby et aL 1985). New
2 :
Mexico locust holds promise as a native southwe~tem counterpart to black locust in
mine reclamation, but it has been used infrequently to date.
Natural invasion and succession occur slowly on most mine sites (Monsen
1984). While New Mexico locust often colonizes sites naturally, there is no
assurance a seed will reach a particular site and establish in a reasonable time frame.
The challenge is to introduce New Mexico locust as mother plants to facilitate
colonization. Seedling production for outplanting on disturbed sites has been
hampered by poor germination (Lin et al. 1996). The goal, then, is to develop
techniques to overcome poor germination in New, Mexico locust, thereby facilitating
macropropagation. LITERATURE REVIEW
Seed Dormancy .
Low gennination in New Mexico locust is primarily the result of seed donnancy, the failure of an intact, viable seed to genninate under favorable conditions (Bewley 1997). Seed donnancy also c,an be defined as the inability ofthe embryo to genninate because of some inherent constraint (Bewley and Black 1994).
The constraint may originate within the embryo, referred to as endogenous or embryo donnancy, or from the tissues surrounding the embryo, known as exogenous or coat imposed donnancy (Baskin and Baskin 1998; Leadem 1997; Bewley and Black
1994).
New Mexico locust seed is donnant when the impenneable seed coat does not allow movement ofwater to the embryo. This specific fonn ofcoat-imposed donnancy, where the restraining tissue involved is an impenneable seed coat, is known as physical donnancy (Baskin and Baskin 1998; Bewley and Black 1994).
Physical donnancy is relatively uncommon in seeds; less than 10% ofall woody species exhibit this condition (USDA Forest Service 1948). However, it is characteristic of many genera of the Fabaceae (bean) family. For example, all successional trees in the genera Gleditsia and Robinia (of which New Mexico locust is a member) produce seeds that exhibit physical donnancy (Baskin and Baskin
4 Physical Dormancy
Seeds that exhibit physical dormancy are said to be "hard seeded," referring to
the thick impermeable seed coat (Leadem 1997). The first line of protection ofthe
testa in leguminous seeds is the outermost waxy cuticle, followed by a subcuticle
consisting ofsuberin. Next is a palisade layer, consisting ofsclereid cells that have
thick, lignified secondary walls. The most common type of sclereid cell in palisade
layers is the macrosclereid. The macrosclereids contain water-repellent substances
including cutin, lignin, suberin and wax (Rolston 1978). This palisade layer is the
primary barrier to water uptake (Serrato-Valenti et al. 1993; Tran and Cavanagh
1984).
All seed coat openings must be blocked to effectively prevent water uptake.
The natural opening in a seed, where it was attached to the mother plant, is the weak
link where physical, dormancy is usually broken (Hopkinson 1993; Kuo and Tam
1988). Anatomical structures associated with the seed coat opening vary from
family to family. In the Papilionoideae, the subfamily ofthe Fabaceae family
containing New Mexico locust, the site ofwater entry is the strophiole. The
:strophiole is a swelling on the seed coat close to. the stalk scar (Hopkinson 1997).
Kelly and van Staden (1987) refer to the strophiole site as the lens, where
macrosclereid cells must separate and raise before water can pass through. Since
·6germination ofdormant seeds ofthe Papilionoideae takes place when the strophiole
permeable (Hopkinson 1993; Kuo and Tam 1988), a logical approach to
5 genninating New Mexico locust artificially would be to loosen the strophiolar cells
and allow water to pass through to the embryo.
Evolutionary Role of Physical Dormancy
Why is physical dormancy, a mechanism that blocks germination, ecologically advantageous? When a plant produc~s seeds with different degrees of dormancy, seeds that are dispersed at one time will germinate over a range oftime
(Bewley and Black 1994). Through variability of dormancy within a seed crop, seeds are exposed to changing environmental conditions, improving the overall chances of survival ofthe seed line (Bewley and Black 1994). Varying expression of dormancy is under genetic control, which indicates that this may be an evolved survival strategy
(Kigel and Galili 1995). A range of dormancy levels may be a particular advantage in the southwestern US, where extreme environmental variations exist.
Methods for Breaking Dormancy
There are three common approaches to breaking dormancy in New Mexico locust seed: acid scarification, hot water soaks and mechanical treatments. In each case, the objective ofthe treatment is to scarify, or abrade, the seed coat to make it permeable to water.
Acid Scarification
Acid scarification consists of soaking seeds in concentrated sulfuric acid, rinsing thoroughly with water upon removal, followed by subsequent drying before
PlaJlltmg. Olson (1974) reports that this scarification method produces the highest
''''·~~j'llal.!Ull results for all Robinia species. Most literature refers to a seminal 1937
6 study conducted by H.G. Meginnis. This stud~, t~ough conducted on black locust seed, is regularly cited in the suggested use of acid to treat New Mexico locust seed as well (Olson 1974; Vines 1960).
Thee published studies document the specific use ofsulfuric acid to treat
New Mexico locust seed. Cox and Klett (1984) report that soaking seeds in concentrated sulfuric acid for thirty minutes improved germination to 34% over an untreated control of4%. Lin et al. (1996) found that, for two seed lots, acid scarification for twenty-five to eighty minutes produced germination of80%. A third lot, however, resulted in less than 20% germination at all durations of 10 to 80 minutes. Hine et al. (1997) showed that acid treatment levels of 1, 2, 5 and 10 minutes resulted in less than 20% germination across five sources. However, there was an increasing trend in germination with acid soak duration and acid treatments were not imposed for as long a time as the previous two studies (30 to 80 minutes).
The mixed results obtained by these investigators were probably due to the widely varying germination response of seed lots to acid treatment. For this reason,
Olson (1974) recommends correlating duration of acid treatment to each seed lot. If treatment duration is too long, the embryo will be permanently damaged. Ifsoak duration is too short, the seed will remain impermeable to water and fail to germinate.
Compounding the problem oftreatment duration is the negative effect acid
have on seedling growth. Any type of scarification can injure seeds to some
by disrupting cells essential for growth, leading to fungal invasion and
·-u..u.u~w injury (Copeland and MacDonald 1995). Some legume seeds subjected
7 to 60 minutes of sulfuric acid scarification (to achieve maximum gennination results) showed more than 50% seedling mortality after gennination (Kelly and van Staden
1987). Bewley and Black (1994) noted that fungal growth on seeds is an indication
of leakage of sugars, organic acids, amino acids and proteins.
Perhaps the greatest drawback to acid scarification is the dangerous nature of
the treatment itself. Difficulties associated with storage and trained use of
concentrated acid may discourage many propagators from using this option (Dreesen
and Harrington 1997). Consequently, hot water and to a lesser extent mechanical
scarification are more commonly used and practical treatments for legume seeds.
Most southwestern container growers use either of these two treatments to scarify
New Mexico locust seed in particular (Hine et al. 1997).
Hot Water Scarification
The widespread use ofhot water scarification to treat locust seed is probably
due to its ease of application and efficacy in comparison to other scarification
treatments. Although this practice had been in use for some time, Wilson (1944) was
one ofthe first to report on its systematic application for black locust. The treatment
consists of placing seeds in boiling water (4 parts water to· 1 part seed by volume),
removing the vessel from the heat source, and allowing the water to cool and seeds to
soak for 24 hours.
Hine et al. (1997), Lin et al. (1996) and Cox and Klett (1984) have examined
. the specific application ofhot water scarification on New Mexico locust seed. Each
these authors compared hot water to acid scarification. Lin et aL (1996) reported
8 60% gennination for two hot water-treated seed lots, while a third lo.t germinated around 50% (as reported above, acid scarification germination ranged from 20% to
80% for these lots). It should be noted that the authors only soaked the seeds in water from 10 to 80 minutes for this experiment, as opposed to the standard 24 hours. Cox and Klett (1984) stated that using the standard 24-hour hot water treatment improved germination over an acid pre-treatment (which produced less than 20% germination), but they did not give specific gennination numbers for the hot water treatment. Hine et al. (1997) found that, across five seed lots, germination was 66% for the standard hot water treatment (as compared to less than 20% germination for acid scarification).
Variations in treatment intensity and duration and their subsequent effects on seed lot
appear to contribute to conflicting conclusions as to whether acid scarification or hot
water treatment is a more successful treatment.
Given mixed results, ease ofapplication is an important argument for use of
hot water application over acid soaks. Safety concerns and material costs for hot
water treatments are low (Dreesen and Harrington 1997). An extension ofthis is the
practicality ofhot water for treating large batches ofseed, where a relative increase in
water volume is all that is needed. However, incr,eases in water and seed, even when
inthe same 4: 1 proportion, can result in dramatically different rates of cooling than in
atest tube experiment. Different cooling rates can lead to contrasting results for the
.. treatment temperature, duration and seed lot (USDA Forest Service 1948).
As with acid scarification, parameters ofhot water treatment application may
to be optimized to maximize germination. Hine et al. (1997) found that as
9 temperature was increased from 70°C to 100°C, overall germination in New Mexico locust seed increased. The 100°C treatment maximized germination for each of five sources tested, with germination usually above 60%.
Prolonged exposure of seeds ofmany woody legumes to boiling temperatures
also may improve germination (Briscoe 1996). This approach has not been tested on
New Mexico locust seed. Exposing seed to prolonged boiling temperatures may
result in fungal infection and subsequent seedling mortality. Marunda (1990) noted
severe fungal infections on both acid and hot water treated Acacia holoserica Cunn.
ex G. Don seeds, attributing the infections to seed damage resulting from the
treatments. Identifying an optimal hot water treatment level that does not result in a
high rate of seedling mortality is a logical goal.
With hot water treatments, as with acid, there is also a danger of overtreating
Seeds and reducing germination (Gosling et al. 1995). The legume Leucaena
leucocephala (Lam.) de Wit germinated equally well after 7 seconds of exposure to
water at 100°C, 7 seconds at 90°C, 30 seconds at 80°C, and 4 minutes at 70°C. In
addition, acceptably high germination was obtained from treatment at 70°C from 3
minutes to 1 hour, whereas no seeds germinated following treatment at 100°C for 2
minutes. Perhaps, along with a greater margin of error for treatment duration, treating
'~ew Mexico locust seeds at a lower temperature may result in a comparatively lower
of seedling mortality. In practice, however, nursery growers tend to use the
and quickest treatment available. The short treatment length at near-boiling
whether recommended or not, may be what is used most frequently.
10 Fine-tuning a treatment in this range oftreatment temperature and duration may be the most logical approach.
Mechanical Scarification
This treatment involves scratching the seed coat with an abrasive material
(USDA Forest Service 1948). One system employs a rotating drum lined with
sandpaper. A paddle spinning at a given number ofrevolutions per minute throws the
seed against the sandpaper (Hine et al. 1997). A second method, dry tumbling,
tumbles seed in a mixer with pea gravel or sharp sand for 5 to 7 days (Dreesen 2001).
These approaches have the advantage that control oftemperature is not required as in
water treatments, nor is safety an issue, as with add scarification.
Hine et al. (1997) mechanically treated New Mexico locust seed using 100
grit sandpaper in a Forsberg (rotating drum) scarifier. Gennination was less than
10%. They concluded that the seed coats were too thin for their apparatus, and
proposed that a finer grit sandpaper would produce more even abrasion. Although
germination varied by seed lot, Dreesen (2001) reported that using a slow-speed dry
tumbling technique for 5 to 7 days produced gennination rates as high as 90%.
As with acid and hot water treatments, treatment duration must be specifically
, correlated to seed lot. Treated seeds are also susceptible to fungal infection
i{Hopkinson 1997). In addition, equipment is specialized and may not always be
'degrading ofstrophiolar cells due to acid and hot water scarification (Miklas et al. 11 1987; Dell 1980; Hanna 1984), but the effect ofmechanical scarification on the strophiole has not been studied. Alternative Method for Breaking Dormancy: Percussion Strophiole degradation has been examined in depth using an impaction treatment, more commonly known as percussion. Hamly (1932) first proposed that shaking or impacting seeds could split the macrosc1ereid cells at the strophiole. Hamly successfully released tension ofbent strophiolar cells ofimpermeable Melilotus alba Medik. (White sweet clover) seeds by delivering physical blows, resulting in cells that had split apart and were only slightly bent. Ninety-two percent oftreated seeds became permeable, as compared to 0.5% of control seeds. Barton (1947) showed that percussion significantly increases permeability in black locust seed. Ballard (1976) reasoned that the integrated nature ofthe strophiole within the seed coat means a blow on any part ofthe seed weakens the strophiole. There are no published studies on the use of percussion on New Mexico locust. Percussion methods vary and often seem SUbjective or ill-defined. Hamly (1932) placed seeds in a glass bottle and shook them three times a second for 20 minutes, resulting in 180 oscillations per minute. Ballard (1973) describes shaking seeds in a closed metal container, with at least fifty seeds per container, at 850-900 \;JS(:~l11alIOIlS per minute for 1 minute. Kelly and van Staden (1987) placed seeds into sealed glass flasks. Flasks were shaken by a wrist-shaker at a rate of 250 ~"'U"'UVJ,~" per minute. Subsequent germination rates were 45% after 2 hours of 55% after 3 hours, 60% after 6 hours, and 100% after 12 hours. 12 There are several potential advantages of percussion over acid, hot water and mechanical scarification. Kelly and van Staden (1987) note that a percussion treatment only alters macrosclereid cells of the strophiole by loosening, separating and raising them. They emphasize that minimal damage is done to the overall anatomical structure of the seed. In acid scarification of the same seed lot, radicle and cotyledon damage occurred. Based on photographic evidence, the authors believe that this morphological damage was due to uneven pressure during imbibition, with water entering haphazardly wherever the acid had sufficiently dissolved the coat. This ultimately led to uneven hydration ofthe embryo. In percussed seeds, water only entered through the s~ophiole when the entire seed coat had been saturated, resulting in controlled and even embryo hydration (Kelly and van Staden 1987). Although all successful scarification treatments may focus at least in part on the strophiole for papilionoid legumes (Baskin and Baskin 1998), percussion may be less damaging to seed coat structures, particularly when compared with acid. Kelly and van Staden did not compare hot water or mechanical scarification methods against percussion. However, the fungal infection associated with hot water and .' mechanical scarification, as with acid scarificatio~, have been directly associated with seed damage (Rehman et al. 1998, USDA Forest Service 1948). Embryo damage in has been observed in hot-water scarified legume seeds (Rehman et al 1998; 13 By limiting embryo damage, percussion scarification reduces seedling mortality compared to other treatments. Kelly and van Staden (1987) noted that seedlings from the percussion treatment were morphologically sound and vigorous when compared to acid treatment seedlings. The authors attribute this improved seedling vigor to the controlled water uptake that limited anatomical damage. Despite a 12-hour percussion treatment, morphological damage was not substantially greater than untreated seeds. In addition, seedling" mass measured 10 days following germination was not significantly different between percussed and untreated seeds. A final advantage ofpercussion may be. the ability to adequately treat the most dormant seeds while avoiding harm to seeds that have already been made permeable by the treatment, or that were permeable to begin with. Instead ofcorroding, scalding or scratching the seed coat, percussion appears to be just a loosening ofthe "strophiole spring" (Ballard 1973). Kelly and van Staden (1987) were able to increase germination with a lengthy treatment (12 hours). Figuring out whether a substantial cushion oftreatment durations exist for percussion ofNew Mexico locust seed would be a significant improvement to current scarification methods, where seed damage and seedling health is an issue. The greatest challenge for applying percussion to New Mexico locust seeds is standardizing a treatment that is both effective and readily reproducible by nurseries. with mechanical scarification, this may require the development ofspecific ~UllDment to properly quantify and execute the treatment. 14 f Objectives The first objective ofthe study is to detennine which scarification treatment results in optimal gennination speed and percentage ofNew Mexico locust. The second objective is to describe treatment interactions with nine seed sources collected in New Mexico. Hypothesis A percussion treatment acts exclusively on the main area ofwater uptake in New Mexico locust seed, loosening the strophiole so that imbibition can occur. This direct mechanism ofdonnancy release will result in improved gennination rate and overall gennination percentage for New Mexico locust seed compared to standard treatments. Standard treatments, including acid, hot water and mechanical scarification, while eventually weakening the strophiolar cleft, cause excessive morphological damage through generalized seed coat" degradation. These treatments are therefore prone to inconsistency due to over and under-treatment. 15 MATERIALS AND METHODS Seed Collection and Cleaning Seeds were collected from August 25 through September 15, 2000, from nine locations along a latitudinal gradient in New Mexico. Seed source locations and descriptive information are given in Figure 1 and Tables I and 2. Stands were separated by at least 45 km. Pods were hand picked as they began to ripen and dehisce (open). Collections were made from ground level to approximately 3 m from several trees at each site. Pods were dried in open-ended brown paper bags for two to four weeks at room temperature (21-230C) to facilitate cleaning. Dried pods were placed in a closed cloth sack, which was pounded against the ground to separate seeds from pods. For pods still containing seeds, hand maceration (crunching pods with hands) completed seed separation. Debris was removed by threshing, with the aid of a greenhouse fan. A South Dakota Seed Blower (Seedbureau, Chicago, Illinois) was used to remove '. remaining chaff and light seeds by air density separation. Seeds were placed in dry storage in sealed quart-size zip-lock bags within a sealed plastic container. Storage temperatures ranged from 2-4oC Seeds remained in storage for approximately 7 months before the experiment. Experimental Design and Treatment Structure The experiment compared the impact ofhot water bath and percussion '''....,l,11'''aLl\./11 on germination properties ofNew Mexico locust seed. A single 16 ' . ...... -.l Figure 1. Seed source collection map. All sources were collected within New Mexico, USA, separated by at least 4S lan. .IIIRZltItW lIJPI'ir· II rllll1l••'_...... ~""'"~~,'_ Table 1. Latitude, longitude, location, elevation and collection date ofNew Mexico locust seed sources. Lot title Latitude (N) Longitude (W) Location Elevation (m) Collection Date Raton 36°58'46" 104°28'72" Raton Pass, NM 2,295 8/25/00 Holman 36°03'06" 105°24'29" Holman,NM 2,476 9115/00 Jemez 35°45'00" 106°41'00" Jemez Springs, NM 1,976 9/9/00 ...... Sandia 35°10'83" 106°23'41" Cedar Crest, NM 2,432 8/26/00 00 Manzano 34°36'89" 106°24'94" Manzano,NM 2,465 9/2/00 Magdalena 33°59'53" 107°09'03" Magdalena; NM 2,544 8/29/00 Capitan 33°27'28" 105°43'25" Alto, NM 2,189 9112/00 Sacramento 32°57'98" 105°44'96" Cloudcroft, NM 2,630 9/14/00 Gila 32°54'00" 107°46'00" Kingston, NM 2,432 9/7/00 f Table 2. Mean seed weight and standard error by source. a Source Mean seed weight (g) for Standard error 100 seeds, n = 5 Raton 2.872 b c 0.0103 Holman 2.522 d 0.0403 Jemez 1.826 e '. 0.0276 Sandia 2.753 c 0.0301 Manzano 2.413 d 0.0370 Magdalena 2.559 d 0.0374 Capitan 2.443 d 0.0458 Sacramento 3.130 a 0.0279 Gila 3.029 ab 0.0348 a Mean seed weights followed by the same letter are not significantly different. 19 experiment was conducted, utilizing a randomized complete block design, blocked by time, with a factorial treatment structure. Experimental factors were seed source and scarification treatment. Nine seed sources from throughout New Mexico were used (Figure 1, Table 1). Scarification treatments included a control, three levels ofhot water and four levels ofpercussion. Hot water treatments included immersion at 100°C, and boiling for periods of 30 seconds and 60 seconds. Percussion treatment levels were 1, 2, 4, and 8 minutes. One hundred seeds per block were used to test each treatment combination. Four blocks of seed 'were tested one at a time (all seed sources included in each run) in close chronological sequence (within 6 weeks). A total of four 100-seed samples were used to test each source by scarification treatment combination. Treatments Hot water treatment ofseeds utilized 50 ml test tubes filled with 30 ml distilled water. Each test tube was used to treat one 100-seed sample at a time, resulting in a ratio of4 parts water to 1 part seed by volume. Test tubes were placed in a beaker, with the beaker filled with tap water to the same level as the water within .,test tubes. Nine test tubes (one per seed source) were placed in the beaker simultaneously, so that one replication ofeach seed source/scarification treatment cotnblnatlon was conducted simultaneously. Hot water treatments include immersion boiling. For immersion treatments, seeds were added to test tubes with the water (boiling), and the test tubes were immediately removed from the beaker to For the boiling treatments, test tubes remained in the beaker with water at 20 boiling temperature for a set duration after the addition ofseeds. Following all treatments, test tubes with seeds cooled 24 hours in a test tube rack. Percussion treatment utilized a pneumatic paint shaker (Central Pneumatic, model #00422, Camarillo, CA). Compressed air was maintained at 530 kPa ±30 kPa (80 psi ± 5 psi), resulting in approximately 350 oscillations per minute. Seeds were placed in 118 ml (4 oz) soil tin cans (US Can, Lobard, IL), and the lids were secured with duct tape. Each can was used to treat one 100-seed sample. To take advantage ofgreatest lateral movement ofthe paint shaker, a spacer (paint can) was used to position the soil tin at the end ofthe shaking arm. This allowed shaking to occur at the greatest distance from the pivot point (fulcrum). Tins were placed perpendicular to the direction ofshaking motion, allowing seeds to impact at greatest force by hitting flat surfaces ofthe soil tin. Germination Testing Before gennination testing, all seeds were soaked 24 hours in reverse . osmosis-treated water (this was included in the process ofhot water treatment j:U~:SCIlOe:a above). Following soaking, seeds were placed in 10.0 cm Petri dishes lined 9.0 cm Whatman qualitative #1 filter paper circles (MG Scientific, Pleasant .. WI). One Petri dish was used for each treatment by source combination. papers were initially moistened with 5 mL reverse osmosis-treated water, then :,\~V'U1.""CU.lJl moistened as needed to ensure adequate conditions for gennination . . were monitored daily (at 3:00 PM) for gennination for 14 days, in accordance .~UI''''Ul'nJlVUa.! Seed Testing Association (ISTA) standards for Robinia species 21 (1STA 1999). Tests were carried out at room temperature (22°C ±O.SoC). Germination was classified as protrusion ofthe radicle from the embryo by at least 1mm, as viewed with the naked eye. X-ray Radiography On August 13,2001, three lOO-seed samples from the Sacramento seed source were treated for 1,4 or 8-minute percussion durations. The seeds were sent immediately by overnight delivery to the Faxitron X-Ray Corporation in Wheeling, Illinois, where Dr. Bob Karrfalt ofthe USDA Forest Service oversaw their imaging. Radiographs were made at SX with a Faxitron Specimen Radiography System Model MX-20. Scanning Electron Microscope Photography On February 19,2002, a Hitachi S-3200N Variable Pressure Scanning Electron Microscope (Mountain View, CA) was used to examine strophiolar cells of untreated seeds and seeds percussed for 4 minutes. Seeds were cut longitudinally ,::....." ....i'> their entire length using a razor blade, splitting the strophiolar region roughly in Cross sections ofseed samples, magnifying the strophiolar region, were viewed 600X under the microscope. Analysis Categorical Analysis of Variance was used to determine treatment, source and (Le., time) main effects and interactions. The response variable was total ...... ~...... v,l.I.. This procedure is a Chi-square test ofhomogeneity using the natural log of germinated to non-germinated seed for each treatment. Maximum 22 likelihood analysis was used to calculate Chi-square test statistics with their observed significance levels. A single contrast was conducted to cempare treatment means against the control. Observed significance levels less than or equal to 0.05 were considered significant. Pairwise comparisons were conducted between non-control treatment mean percents, using a conservative alp~a value of 0.05 divided by the number of comparisons (Bonferroni approach) and approximate pairwise z statistics. Categorical analyses were performed using Proc Catmod (SAS Institute 1989). RESULTS Scarification, seed source, block and all interactions between factors influenced germination (Table 3). The main effects are reported in the presence of these interactions because the overall robustness oftreatments and seed sources are of interest. "Background" levels ofnaturally occurring variation in seed sources are seen in control germination percentages in Figure 2. In particular, "time" variability is unlikely to be under the control ofthe producer. Appendix B details experimental data for source by treatment gennination percents for each ofthe four blocks. Averaged over all treatments, germination by source ranged from 40% to 79% (Figure 2). Averaged over all sources (Figure 3), and for each source (Figure 4, 5), percussion scarification durations of4 or 8 minutes yielded greatest germination .. Averaged over all sources, germination increased with increasing durations of percussion, then leveled offfor the 4- and 8-minute treatments (Figure 3). Immersion and 30-second boiling scarification maximized germination for the hot water treatments, with a decrease in germination when seeds were boiled for 60 seconds For all but two sources, there is a similar ipcreasing trend in germination from 1to 4 minutes ofpercussion (Figure 4). The Holman source increased in germination 1 to 2 minutes, then leveled off at 4 minutes. The Jemez source changed little 1 to 4 minutes oftreatment. As treatment duration increased from 4 to 8 !mutes. five sources decreased in germination (Figure 4a), whereas three sources 24 Table 3. Categorical Analysis of Variance table for germination response to seed treatment, seed source, block and the interaction of the factors. Source of DF Chi-Square Observed Variability Significance Level Source 8 983.70 <0.0001 Trt 7 2338.76 <0.0001 contrast 1 169.76 <0.0001 Source*Trt 56 ,970.77 <0.0001 Block 3 55.38 <0.0001 Source*Block 24 101.62 <0.0001 Trt*Block 21 75.64 <0.0001 Source*Trt*Block 168 342.83 <0.0001 r -] Northern NM Sources Central NM Sources 80 Southern NM Sources 4) S 8 60 -I-.-" ...... "." ...... - ...... "...... ~..- ...·····"1 ~ 0... § '.g ~ 40 .~ N 0 0'\ 20 o 'f..0<> ~ ~4 'jy'b ((}i'<>o ';:,.~~ .K/1<> :§-o 6~'b- ~ ~# \~~ ~~ 0'b-~ ~ ~'b-~ ~'b-~'b' ~'b-v(!i' SOURCE Figure 2. Effect ofNew Mexico locust seed source on germination for data averaged over all scarification treatments. Control germination is shown by solid line within bar. 80 ~ cd ~ o lmmersion Boiling 30 sec Boiling 60 sec Control 1 min 2min 4 min 8 min Hot Water Treatments Percussion Treatments Figure 3. Effect ofscarification on germination ofNew Mexico locust for data averaged over all seed sources. Means labelled with the same letter are not significantly different. 100~------~--~------~ a I ..··· .. ·······.. ·· ..·:I······.... . ' :E...... _"...~""iii;:.:...-::::. •. _ .. _ ...... •..... f ,."...... --...~--...... /' ~ ...... "--."-- ... 80 ;t-- - - I ..~/' ------...... - ~ ...... - L ...... ---r II...... 60 fi - ...... Holman .. ~ .... ··-I 80 ./ ;j:::::::::~:=:~=-~~-:::'1 ." ./ ··' .. /// ./ ./ .I // ~ • .' / .,r ./:t'/ ././ 60 :t.r ./ // /'Y ',-.. -...-..-..-..-.-M-a-gd-a-Ie-n-a--' 1: /' /' _.. _.. Capitan . ---- Manzano 40 f:. ~y -_. Raton 2 4 8 Time interval ofpercussion scarification (min) Figure 4. Effect of seed source by treatment interaction on New Mexico locust germination for percussion scarification, pattern a and pattern b. 28 TIT 80 -...... ::-::,.;.....--... . - . - . -;...-:.:"="'..-., - -'-:::"'-".cE= . ..-----.:.:.:-...:.:=..·f,··-··-··-·· -" ...... ,.:...._-- ...... -::::... --- .'.~ ...... *~... -- ...... ,--"""", 1 - ...... ~ ...... ':"'-~- ~ t ...... _..:..::.:.:,...... d 60 ...... ~ -- A.. ------:r...... 1 s:: ...... o ...... I ...... Raton :1 40 Holman tv \0 -_. Jemez c3 --- Sandia Manzano 20 Magdalena Capitan Sacramento Gila o Immersion Boiling 30 sec Boiling 60 sec 'Hot Water Scarification Figure 5. Effect of seed source by treatment interaction on New Mexico locust germination for hot water scarification. continued to increase in germination (Figure 4b). There was no difference between the two treatment levels for the Manzano source (Figure 4b). The effect ofhot water scarification also varied among sources, with few overall differences between hot water treatments (Figure 5). However, boiling for 60 seconds yielded highest germination for the Jemez source, whereas increasing durations ofboiling decreased germination for the Raton source. In addition to consistently high germination, 4- and 8-minute percussion treated seeds germinated as fast or faster than all other treatments (Figure 6, Table 4). 050, measured as the number of days needed to complete 50% ofeventual germination, averaged 2 days for both ofthese treatments. X-ray radiographs were taken of seeds collected from the Sacramento source percussed for 1, 4, or 8 minutes (Figure 7). Seeds show signs of damage at 8 minutes ofpercussion '(Figure 8). Compared to an undamaged seed percussed for only 4 minutes (Figure 8a), damage at the 8-minute level includes fractured radicles, pitting of the seed coat and chips in seed coat structure (Figure 8b-d). Scanning electron microscope (SEM) images at 600X reveal the direct effect . of 4 minutes of percussion on cells ofthe strophiolar region (Figure 9). Whereas untreated seeds retain cellular organization in the strophiolar region, strophiolar cells seeds treated for 4 minutes ofpercussion separate and raise, resulting in a ~-~"e:."""'~U"""" appearance. 30 U·-~·--l-~---·.y.···-----·········--·--········-·······-...... - .. -e-- Percussion 8 min -0- Percussion 4 min ---T- Percussion 2 min ----- 2 4 6 8 10 12 14 Day Figure 6. Timecourse ofNew Mexico locust gennination following scarification treatments. G50 is the number ofdays to 50% of eventual germination. 31 Table 4. Effect ofseed source and treatment on G50 (days to 50% germination). Seed Source Control Immersion B30sec B60sec Plmin P2min P4min P8min. mean Raton 3 3 5 7 4 3 2 2 4 Holman 2 2 2 2 2 2 2 2 2 Jemez 4 6 6 5 3 2 2 2 4 Sandia 2 2 2 2 2 2 2 2 2 w tv Manzano 5 2 2 3 3 2 2 2 3 Magdalena 4 2 2 3 4 3 2 1 3 Capitan 5 4 2 2 3 2 2 2 :3 Sacramento 3 2 2 2 4 3 2 2 3 Gila 3 2 2 2 2 2 2 2 2 mean 3 3 3 3 3 2 2 2 3 standard 1.1 1.4 1.6 1.8 0.9 0.5 0 0.3 0.8 deviation Note. Columns and rows may not agree due to rounding. Figure 7. X-ray radiographs ofSacramento seed source percussed for 1, 4 or 8 minutes. 33 Figure 8a-d. Seed damage close-ups. Figure a shows undamaged seed percussed for 4 minutes. Figures b-d show seeds percussed for 8 minutes. Damage includes fracturing ofradicle in b, complete removal ofseed coat in c and pitting ofseed coat in d. 34 Untreated Treated Figure 9. SEM images ofuntreated vs. 4-minute percussion-treated strophiolar cells ofNew Mexico locust seeds. Treated cells are loosened, separated and raised. DISCUSSION Determining the Optimal Treatment Averaging results of scarification treatmems across seed sources is a useful tool to predict which treatment is most likely to be successful for an unknown seed source. Averaged over all sources, four- and eight-minute percussion treatments resulted in the highest germination, 81.5% and 79.0% respectively, and produced germination that was as fast or faster than all other treatments. Although there was no difference between these two treatments on average, an argument for 4 minutes of percussion as a standard protocol for treating New Mexico locust seeds can be made. Descriptively, 4 minutes ofpercussion improved germination over 8 minutes ofpercussion for a majority of sources (five out ofnine), 8 minutes of percussion improved germination for three ofnine sources, with no difference for a fmal source. For individual sources, 4 minutes ofpercussion was always the best or second· best treatment and was always equal to or better than all hot water treatments. Germination at 8 minutes ofpercussion was lower than all other percussion durations for the Jemez and Gila sources and was lower than at least one hot water treatment for these sources as well. The x-ray radiographs ofthe Sacramento seed source illustrate the excessive internal and external seed damage that can occur with 8 minutes ofpercussion. germination counts, 8-minute percussed seeds from Holman and Capitan ;,."ClUll"'" signs of damage that were visible to the naked eye (see below). r~everthe:les:s. some ofthese seeds still fit the criteria for germination. Neither x-ray 36 photography nor visible observations during seed counts revealed seed damage at the 4-minute percussion level. In operational trials, seeds percussed for 4 minutes developed into healthy seedlings (Khadduri and Harrington, unpublished data). Strophiole and Seed Coat Behavior Four minutes ofpercussion successfully breaks seed dormancy in a large percentage ofNew Mexico locust seeds without excessively damaging the surrounding seed coat or internal seed coat structures. Most papilionoid legume seeds, including seeds ofNew Mexico locust, have a specific region of the seed coat known as the strophiole or lens (Hamly 1932). The strophiole is located on the cotyledonary lobe ofa locust seed (Figure 10 a-c). Ifseeds are not indiscriminately damaged, the strophiole is the first point ofwater entry to the embryo (Hopkinson 1997). When a papilionoid seed is percussed for the appropriate amount of time, repeated hits on the integrated seed coat loosen the constrained cells ofthe strophiolar region without excessively damaging the rest ofthe seed coat (Ballard 1976). SEM photos illuStrate that 4 minutes ofpercussion loos,ened and separated strophiolar cells ofNew Mexico locust seeds when compared with constrained, organized strophiolar cells ofuntreated seeds (Figure 9). Research-has shown that water passes to the embryo through the treated strophiole in a controlled manner and regulated entry of water to the embryo is associated with even pressure on underlying seed tissues .(Kelly and van Staden 1987). 37 10. Strophiole ofunimbibed New Mexico locust seed (a) and following uptake (b,c). 38 In contrast, hot-water treatments can excessively degrade the seed coat (Rebman et al. 1998, Marunda 1990). Random cracks in the seed coat can promote irregular water uptake associated with uneven pressure on underlying seed tissues and subsequent seed damage (Rebman et aI. 1998, Marunda 1990). Along with a proportionately large amount ofun-imbibed (thus, under-treated) seeds (Figure l1a), damage was observed in hot-water treated seeds from six sources during germination counts. Damage included seeds where the radicle failed to elongate (Figure 11 b), seeds where the radicle elongated but was delayed in freeing itselffrom the seed coat (Figure 11c), and seeds where the radicle completely broke off from the embryo (Figure lId). In addition, some seeds germinated in reverse (the cotyledon emerged fIrst). Ifthe radicle emerged completely from the seed coat, they were counted as germinated. While nearly all seeds treated for 8 minutes ofpercussion imbibed water, damage was evident in x-ray evaluations (Sacramento seed source) and in germination counts (Holman and Capitan) at this treatment level. Seed damage during germination counts was similar to levels found in hot-water treated seeds, suggesting that over-treatment by both hot water and longer durations ofpercussion may result in similar damage to seed structure. The Sacramento and Holman seed sources were two ofthe fIve sources that op.c:rp!'l the benefIts ofloosening the strophiole, as evidenced by imbibition of all 8-minute treated seeds, were outweighed by external and internal seed coat 39 a b c d 11. Damage to New Mexico locust seeds observed during germination cotmts following hot water scarification. 9a. Hot water replicate, including ~...t;,""" seedlings and numerous unimbibed seeds. 9b. Close-up of seed where raallCle failed to elongate. 9c. Close-up ofseed where radicle elongated but was '-':.>I'"..n in freeing itself from the seed coat. 9d. Close-up ofseed where radicle broke offfrom embryo. 40 damage (which also may be contributing to high imbibition). For the three sources for which 8 minutes ofpercussion resulted in maximum germination, extended percussion of seeds necessary to loosen the strophiole may not have excessively damaged the seed anatomy. The Manzano source did not significantly change in germination from 4 to 8 minutes ofpercussion, suggesting that an increase in permeable seeds at the 8-minute treatment duration was balanced by the number of seeds that failed to germinate due to seed damage. To summarize, the 4-minute percussion treatment improved germination over 8 minutes ofpercussion for a majority ofsources (five ofnine), the 8-minute percussion treatment improved germination over 4-minutes ofpercussion for three sources, and there was no difference with the Manzano source. Notably, for the Raton source, 8 minutes of percussion increased germination by 23.5 percentage points over 4-minutes of percussion (78.75% vs. 55.25% respectively), the next best treatment. Raton was the only source to result in germination less than 75% for 4 minutes ofpercussion, so a grower may wish to apply longer percussion durations to lots that do not respond well to this treatment level. Finally, there was an interesting trend in the Jemez and Raton sources, seed that significantly contributed to the source by treatment interaction (Figure 12). A ~"'...u" test showed that the Jemez source had a significantly lighter seed weight than other sources, while the Raton source had an average seed weight. For the Jemez the longest duration of boiling increased germination. Conversely, only short ,,"U4UVll;:) ofpercussion were needed to achieve high germination and the longest 41 80 _ Jemez a .. Raton 20 0-'----- Immersion Boiling 30 sec Boiling 60 sec Hot Water Scarification 10°Tr======~------~------' _ Jemez b .. Raton 80 iD .....~ C iD 60 eiD Q.. c 0 .~ c 40 .~ iD 0 20 0-'----- 1 min 2 min 4 min 8 min Percussion Scarification Figure 12. Effect of hot water scarification (a) and percussion scarification (b) on interaction ofJemez and Raton seed sources. 42 --~----~~~~======~----...... duration ofpercussion reduced germination. For the Raton source, longer durations ofboiling consistently decreased germination, whereas longer percussion treatments consistently increased germination. A possible explanation is that a thick seed coat is associated with a loose strophiole structure, and that a thin seed coat is associated with a constrained strophiole. This contrast merits further study and may suggest an ecological adaptation. Further investigation ofthis phenomenon may contribute to successful pre-screening ofunknown seed sources to determine optimal treatment leveL 43 CONCLUSION Based on results obtained in pilot study (see Appendix) and thesis experiments, 4 minutes ofpercussion is recommended as a standard protocol for treating an unknown source ofNew Mexico locust seed. In addition to laboratory results, the 4-minute percussion treatment has resulted in germination greater than 80% and healthy seedling grow-outs for production purposes (unpublished data). In laboratory experiments, only one hundred seeds per 118 ml (40z) soil tin were percussed at a time, whereas for production we have treated up to five hundred seeds per tin. For large-scale production ofNew Mexico locust seedlings, growers may wish to experiment with larger quantities of seed and larger-size soil tins. Seed storage trials following percussion treatments were not conducted. However, germination greater than 80% and healthy seedlings were obtained when seeds were percussion-treated and subsequently stored at room-temperature (22°C) for one week (unpublished data). Seed storage trials merit further study, as the standard hot water treatment leaves seeds in a softened condition that is not conducive to storage. Hydroseeding is an effective method for direct seeding in reclamation situations. Percussion treatment may be conducive to hydro seeding, as seeds remain a dry, relatively undamaged, yet permeable condition following treatment. Sulfuric seeds also remain dry following treatment, but extensive damage to seed coat poor seedling vigor have been reported with this treatment (Kelly and van Staden . A grow-out study using a hydro seeder on a reclamation site testing percussion 44 against other treatments, such as sulfuric acid and mechanical scarification, also " . merits further study. Papilionoids are the largest subfamily oflegumes, covering almost all legumes occurring in temperate climates (Baskin and Baskin 1998). As with New Mexico locust, these legumes often play an integral role in the revegetation of disturbed lands. The pilot study (Appendix) indicates that percussion scarification successfully treats the widely used black locust as well. Developing superior scarification methods to the standard hot water, mechanical scarification and acid treatments may facilitate the use ofpapilionoid legumes in reclamation projects. 45 APPENDICES APPENDIX A PILOT STUDY: PERCUSSION AS AN ALTERNATIVE SCARIFICATION FOR NEW MEXICO LOCUST AND BLACK LOCUST SEEDS 2 4 Nabil Y. Khadduri', John T. Harrington , Lee S. Rosnet and David R. Dreesen Hot water and sulfuric acid soaks are traditional treatments for seeds of many temperate woody legwnes, including locusts. However, these scarification techniques often produce inconsistent germination. Percussion scarification, where seeds are repeatedly propelled against a hard surface, was compared with hot water scarification to evaluate treatment efficacy for New Mexico locust (Robinia neomexicana) and black locust CR. pseudoacacia). seeds. In the hot water treatment, seeds were placed in a 98°C water bath, which was immediately removed from the heat source. For percussion scarification, seeds were placed in a soil sample tin and agitated in a paint shaker for 1, 2, 3,4, 5 or 10 minutes. All treatments, including the control, were followed by 24-hour water soaks. Hot water baths resulted in 56% and 41 % germination for New Mexico locust and black locust respectively. For both species; nearly all durations ofpercussion increased germination over the hot water . treatment. Percussion durations of 4, 5 and 10 minutes for New Mexico locust and 3, 4 and 5 minutes for black locust resulted in at least 90% germination. Traditional scarification treatments randomly degrade the entire seed coat, which can lead to tissue damage during water uptake. Percussion scarification specifically weakens the strophiole, the natural source ofwater entry to the seed in papilionoid legwnes. INabil Y. Khadduri, Graduate Research Assistant, New Mexico State University, Las Cruces, NM, 88003, nkhadduri@hotmaiLcom .2John T. Harrington, Associate Professor, New Mexico State University Mora Research Center, Mora, NM, 87332, [email protected] 3Lee S. Rosner, Science Specialist, New Mexico State University, Mora, NM 87332, [email protected] 4David R. Dreesen, Agronomist, USDA Natural Resources Conservation Service, Los Lunas Plant Materials Center, Los Lunas, NM 87031 48 Following percussion, imbibition is controlled through the strophiole and underlying tissue is protected. Keywords: .New Mexico locust, black locust, scarification, percussion, germination Introduction Locust trees (Robinia spp.) are aggressive pioneer species that quickly colonize disturbed land, fix atmospheric nitrogen in the, soil, and stabilize slopes that are prone to erosion (Klemmedson 1994, Ashby et al. 1985) (Figure A1a,b). Recognition ofthese attributes has contributed to increased demand for locusts in reclamation projects. Locust seeds, however, exhibit physical dormancy, where a thick seed coat prevents the movement ofwater and gasses to the embryo (Leadem 1997). Sulfuric acid or hot water soaks have traditionally been used to break seed dormancy in macro-propagation of locusts. Many growers are moving away from the use ofsu1furic acid in treating locust seeds (Dreesen and Harrington 1997). In addition to being dangerous, acid soak durations must be specifically correlated to. seed lot (Olson 1974). Collections ofnative seed can vary tremendously in seed size, seed weight and hard seededness, etc. This variability limits the utility of sulfuric . acid soaks, as even within a given lot, some seeds may be damaged, whereas others are under-treated and fail to imbibe (take up water). A survey ofSouthwestern container growers indicates that hot water baths are currently the most common . scarification method for locust seeds (Hine et al. 1997). However, hot water baths also produce inconsistent germination (Lin et al. 1996). 49 b Figure AI. New Mexico locust: a primary invader following disturbance. Ala. Crown-sprouting at edge ofburn near Los Alamos, NM--one year after Cerro Grande fire. Alb. Colonizing road cut, Gila National Forest, NM. 50 Percussion scarification, where seeds are repeatedly propelled against a hard surface, is an alternative dormancy-breaking method that has proven successful in legumes related to locusts (Hamly 1932, Barton 1947). We compared this alternative treatment to hot water scarification to evaluate treatment efficacy for New Mexico locust and black locust seeds. Materials and Methods Commercial New Mexico locust seeds (Western Native Seed, Coaldale, CO) were collected fall 2000 in Huerfano County, CO. Black locust seeds were collected September 2000 from Taos County, NM by harvesting ripe pods from several trees at a distance of up to 3 m from ground level. Black locust seeds were allowed to air-dry for 3 weeks, and were threshed and separated from large chaffusing a greenhouse fan. A South Dakota Seed Blower (Chicago, Illin This study compared hot water scarification with percussion scarification using treatment and species as experimental factors. Along with a control, scarification treatments included immersion ofseed in boiling water as well as percussion durations of 1, 2, 3, 4,5 or 10 minutes. Four 100-seed samples were used to test each species by scarification treatment combination. For the hot water treatment, test tubes were filled with 30 mL water and placed in a water-filled beaker (Figure A2a). The bath was raised to boiling (-98C at 51 b c Figure A2. Implementation ofhot water and percussion scarification. A2a. Hot water bath. A2b. Paint-shaker set-up. A2c. Placing soil tin containing seeds at end ofpaint shaker ann. 52 Las Cruces, NM elevation of 1300 meters). Each 100-seed replicate was placed in a test tube, and each test tube was immediately removed to cool at room temperature. Percussion scarification was implemented using a pneumatic paint shaker (Central Pneumatic, model #00422, Camarillo, CA) (Figure A2b). This allowed standardization oftreatment intensity. Compressed air pressure was maintained at 530kPa ± 30kPa (80psi ± 5psi), resulting in approximately 350 oscillations per minute. Each lOO-seed sample was placed in a 118 ml (4 oz) soil tin for shaking. To maximize lateral movement of paint shaker, a paint can was used as a spacer to place , soil tins at the end ofthe shaking arm (Figure A2c). This allowed shaking to occur at the greatest distance from the pivot point (fulcrum). Following all treatments, including the control, seeds were soaked for twenty-four hours in distilled water prior to plating out. Gennination was evaluated on lab benches at room temperature. Each replicate of seeds was placed in a 10.0 cm petri dish on moistened filter paper, with humidity maintained by enclosing the petri dishes in ziplock bags. Gennination was monitored daily for 14 days, in accordance with International Seed Testing Association standards for Robinia species (ISTA 1999). Gennination was defined as protrusion of the radicle from the embryo by at least 1mm, as viewed by the naked eye. Categorical Analysis of Variance using Prpc Catmod (SAS Institute 1989) was used to detennine treatment differences in germination percentages for each source. This procedure is an extension of a Chi-square test of homogeneity using the 53 natural log ofthe ratio of germinated to non-germinated seeds for each treatment. P values less than 0.05 were considered significant. A limited set ofpairwise comparisons was conducted to compare treatment mean percents; using a conservative alpha value of 0.05 divided by the number of comparisons. Results Percussion maximized germination for both New Mexico locust and black locust. There were three germination response levels for New Mexico locust (Figure A3). First, there was a low response in the control, indicating a dormant seed lot. The hot water treatment and I-minute duration ofpercussion share a second, intermediate response. Germination was highest at percussion levels from 2 to 10 minutes. A selected percussion treatment intermediate in time (5 minutes) produced the steepest germination curve, indicating the fastest germination rate (Figure A4). The hot water treatment produced an intermediate curve, or relatively slower germination rate. The control curve was very flat, indicating the slowest germination rate for New Mexico locust. Control seeds germinated poorly for black locust as well (Figure A5). Hot water scarification improved germination over the control, but all levels ofpercussion improved germination over hot water. The optimal durations ofpercussion treatment were bracketed, with a significant increase from the 1to 2 minute percussion level, a high germination response from 2 to 5 minutes ofpercussion, and finally a significant 54 100~------' 80 ~ d 25 60 CL) p..."'" t:1 .....0 ~ t:1 'g 40 Vl ~ Vl '20 o Control Hot Water Percussion Percussion Percussion Percussion Percussion Percussion I min 2 min 3 min 4 min 5 min 10 min Scarification Figure A3. Effect ofscarification on germination response for New Mexico locust. *=significant difference between adjacent treatment bars, ns = no significant difference. 100·lr::~::~~'------e- Percussion 5min ·.. 0·.. Hot Water -..,.-. Control 80-J:=.:·,=.===== ~ ~ ·~,- 1 .-----.~--,---,------(\) 6b-·-l,······ .."··..··.... ,·..-·,- ..-··- ...... ·... ·, ..-··.. ,· ..··,'-,·" ...... u a:> 0 Il-t ...... 0 ...... 0 ...... ·····0················0 od .~ ~ .~ 40 c3 20 ,0' _---~--- --~-----y-----~------T-----~ -"T o Day."" 0 Day I Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Time Figure A4. Effect of selected scarification treatments on germination speed for New Mexico locust. First seven days of fourteen-day experiment are shown. ns 80 ~ cd 5 60 (,) [) Cl.. d .....0 1t1 .....d § 40 Ul d -...J 20 o Scarification Figure A5. Effect of scarification on germination response for black locust. * = significant difference between adjacent treatment bars, ns = no significant difference. decrease at the 10-minute level. This drop-off suggests that 10 minutes of percussion over-treats seeds from this lot of-black locust. A selected percussion treatment intermediate in time (5 minutes) also had the fastest germination rate for black locust (Figure A6)_ The hot water curve is somewhat flatter, indicating a slower germination rate. The control curve is very flat, indicating a slow germination rate. A qualitative difference between hot water and percussion-treated seeds was also observed. Along with healthy germinants, the hot water treatment produced over-treated seeds (Figure A 7a). Examples ofdamaged germinants include seeds where the radicle did not elongate (Figure A 7b), seeds where the radicle elongated but was delayed in freeing itself from the seed coat (Figure A7c) and seeds where the radicle broke off completely from the embryo (Figure A7d). In addition to these examples of over-treatment, there were numerous small, dark seeds that failed to imbibe water. For percussion treatment, nearly all seeds took up water (Figure A8a). Germination was rapid and uniform, with little damage evident (Figure A8b). To summarize, both New Mexico locust and black locust seed lots were dormant. The standard hot water scarification treatment improved germination, but only to 56% and 41 % for New Mexico locust and black locust, respectively. Nearly all percussion durations improved total germination as well as germination rate over the hot water treatment. Germination was greater than 90% for percussion durations of4, 5 and 10 minutes for New Mexico locust and 3, 4 and 5 minutes for black locust. 58 100r=:=~==~~=r------~ ___ Percussion 5 min ···0··· Hot Water -lIf'-' Control 80r======~--~~------~ CI) s>t:: CI) 60 0 i> 0...... § ttl .....t:: E 40 Vl \0 c3 20 '-~'-~---"~"'--'---'''------'--~'------I o· _~_------lIf'-----~ •...... --------_... .. O I Day 0 Day I Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Time Figure A6. Effect of selected scarification treatments on germination speed for black locust. First seven days of fourteen-day experiment are shown. a b c d Figure A7. Damage to New Mexico locust seeds observed during germination counts following hot water scarification. A7a Hot water replicate, including damaged seedlings and numerous unimbibed seeds. A7b. Close-up ofseed where radicle failed to elongate. A7c. Close-up ofseed where radicle elongated but was delayed in freeing itself from the seed coat. A7d. Close-up ofseed where radicle completely broke offfrom embryo. 60 a b Figure A8. New Mexico locust seeds following percussion treatment at time of germination counts. A8a. Percussion replicate with uniformly healthy seedlings and near complete imbibition of seeds. A8b. Close-up ofpercussion-treated seedling. 61 Discussion Papilionoid seeds have a specific anatomical feature known as the strophiole, or lens-the natural site of water entry to the seed (Hamly 1932). The strophiole is located on the cotyledonary lobe of a locust seed (Figure A9). When a papilionoid legume seed is percussed for the appropriate amount of time, repeated hits on the integrated seed coat loosen the constrained cells ofthe strophiolar region, without excessively damaging the rest ofthe seed coat (Ballard 1976). When a percussed seed is soaked, water enters exclusively through the strophiole in a controlled manner (Kelly and Staden 1987). This regulated entry of water to the embryo is associated with even pressure on underlying seed tissues. This contrasts with seeds that have been hot water or acid scarified, treatments that can randomly degrade the seed coat. Unlocalized cracks in the seed coat can promote irregular water uptake associated with uneven pressure on underlying seed tissues and subsequent seed damage (Rehman et al. 1998, Marunda 1990). Papilionoids are the largest subfamily of legumes, covering almost all legumes occurring in temperate climates (Baskin and Baskin 1998). As with New Mexico locust and black locust, these legumes often play an integral role in the revegetation ofdisturbed lands. Developing superior scarification methods to the standard hot water and acid treatments should fadlitate the use ofthese legumes in restoration projects. 62 Figure A9. Strophiole ofunimbibed New Mexico locust seed (a) and following water uptake (b,c). 63 APPENDIXB GERMINATION PERCENTS AND STANDARD ERRORS FOR SOURCE, BLOCK AND TREATMENT FOR NEW MEXICO LOCUST Table B 1. Gennination percents and standard errors for source, block and treatment for New Mexico locust. Source Block Treatment Germination Percentage Standard Error Capitan 1 Immersion 74 4.37 Capitan 2 Immersion 69 4.62 Capitan 3 Immersion 71 4.54 Capitan 4 Immersion 67 4.70 Capitan Boiling 30 sec 79 4.07 Capitan 2 Boiling 30 sec 76 4.27 Capitan 3 Boiling 30 sec 72 4.49 Capitan 4 Boiling 30 sec 76 4.27 Capitan Boiling 60 sec 71 4.54 Capitan 2 Boiling 60 sec 54 4.98 Capitan 3 Boiling 60 sec 67 4.70 Capitan 4 Boiling 60 sec 76 4.27 Capitan I Control 5 2.18 Capitan 2 Control 7 2.55 Capitan 3 Control 4 1.96 Capitan 4 Control 5 2.18 Capitan 1 Percussion I min 47 4.99 ..,.., Capitan 2 Percussion I min 4.70 .., Capitan Percussion I min ""39 4.88 Capitan 4" Percussion I min 38 4.85 Capitan I Percussion 2 min 55 4.97 Capitan 2 Percussion 2 min 71 4.54 Capitan .. Percussion 2 min 65 4.77 Capitan 4" Percussion 2 min 56 4.96 Capitan I Percussion 4 min 86 3.47 Capitan 2 Percussion 4 min 79 4.07 Capitan 3 Percussion 4 min 81 3.92 Capitan 4 Percussion 4 min 82 3.84 Capitan I Percussion 8 min 90 3.00 Capitan 2 Percussion 8 min 88 3.25 Capitan 3 Percussion 8 min 90 3.00 Capitan 4 Percussion 8 min 90 3.00 Gila I Immersion 71 4.54 Gila 2 Immersion 77 4.21 .., Gila Immersion 75 4.33 Gila 4" Immersion 72 4.49 Gila I Boiling 30 sec 68 4.66 Gila 2 Boiling 30 sec 81 3.92 Gila 3 Boiling 30 sec 54 4.98 Gila 4 Boiling 30 sec 63 4.83 Gila I Boiling 60 sec 55 4.97 Gila 2 Boiling 60 sec 68 4.66 Gila 3 Boiling 60 sec 57 4.95 Gila 4 Boiling 60 sec 70 4.58 65 Table B 1. Continued. Source Block Treatment Germination Percentage Standard Error Gila 2 Control 13 3.36 Gila 3 Control 13 3.36 Gila 4 Control 20 4.00 Gila 1 Percussion 1 min 66 4.74 Gila 2 Percussion 1 min 57 4.95 Gila 3 Percussion 1 min 73 4.44 Gila 4 Percussion 1 min 82 3.84 Gila 1 Percussion 2 min 81 3.92 Gila 2 Percussion 2 min 86 3.47 Gila 3 Percussion 2 min 85 3.57 Gila 4 Percussion 2 min 54 4.98 Gila 1 Percussion 4 min 79 4.07 Gila 2 Percussion 4 min 83 3.76 Gila 3 Percussion 4 min 84 3.67 Gila 4 Percussion 4 min 79 4.07 Gila 1 Percussion 8 min 51 5.00 Gila 2 Percussion 8 min 76 4.27 Gila 3 Percussion 8 min 52 5.00 Gila 4 Percussion 8 min 74 4.39 Holman 1 Immersion 83 3.76 Holman 2 Immersion 86 3.47 Holman 3 Immersion 81 3.92 Holman 4 Immersion 90 3.00 Holman 1 Boiling 30 sec 90 3.00 Holman 2 Boiling 30 sec 85 3.57 Holman 3 Boiling 30 sec 87 3.36 Holman 4 Boiling 30 sec 83 3.76 Holman 1 Boiling 60 sec 79 4.07 Holman .2 Boiling 60 sec 80 4.00 Holman 3 Boiling 60 sec 86 3.47 Holman 4 Boiling 60 sec 93 2.55 Holman Control 25 4.33 Holman 2 Control 19 3.92 Holman .," Control 21 4.07 Holman 4 Control 24 4.27 Holman Percussion 1 min 77 4.21 Holman 2 Percussion 1 min 85 3.57 Holman 3 Percussion 1 min 80 4.00 Holman 4 Percussion 1 min 89 3.13 Holman 1 Percussion 2 min 88 3.25 Holman 2 Percussion 2 min 96 1.96 Holman 3 Percussion 2 min 93 2.55 Holman 4 Percussion 2 min 96 1.96 Holman 1 Percussion 4 min 92 2.71 Holman 2 Percussion 4 min 97 1.71 66 I Table B 1. Continued. Source Block Treatment Germination Percentage Standard Error Holman 3 Percussion 4 min 91 2.86 Holman 4 Percussion 4 min 96 1.96 Holman 1 Percussion 8 min 88 3.25 Holman 2 Percussion 8 min 85 3.57 Holman 3 Percussion 8 min 83 3.76 Holman 4 Percussion 8 min 82 3.84 Jemez 1 Immersion 53 4.99 Jemez 2 Immersion 44 4.96 Jemez 3 Immersion 52 5.00 Jemez 4 Immersion 70 4.58 Jemez 1 Boiling 30 sec 58 4.94 Jemez 2 Boiling 30 sec 55 4.97 Jemez 3 Boiling 30 sec 48 5.00 Jemez 4 Boiling 30 sec 65 4.77 Jemez 1 Boiling 60 sec 68 4.66 Jemez 2 Boiling 60 sec 63 4.83 Jemez 3 Boiling 60 sec 73 4.44 Jemez 4 Boiling 60 sec 89 3.13 Jemez 1 Control 2 1.40 Jemez 2 Control 9 2.86 Jemez 3 Control 6 2.37 Jemez 4 Control 8 2.71 Jemez 1 Percussion 1 min 58 4.94 Jemez 2 Percussion 1 min 63 4.83 Jemez 3 Percussion 1 min 77 4.21 Jemez 4 Percussion 1 min 89 3.13 Jemez 1 Percussion 2 min 68 4.66 Jemez 2 Percussion 2 min 64 4.80 Jemez 3 Percussion 2 min 80 4.00 Jemez 4 Percussion 2 min 83 3.76 Jemez Percussion 4 min 68 4.66 Jemez 2 Percussion 4 min 79 4.07 Jemez 3 Percussion 4 min 70 4.58 Jemez 4 Percussion 4 min 83 3.76 Jemez 1 Percussion 8 min 52 5.00 Jemez 2 Percussion 8 min 59 4.92 Jemez 3 Percussion 8 min 61 4.88 Jemez 4 Percussion 8 min 70 4.58 Magdalena 1 Immersion 69 4.62 Magdalena 2 Immersion 69 4.62 Magdalena "J Immersion 70 4.58 Magdalena 4 Immersion 85 3.57 Magdalena 1 Boiling 30 sec 65 4.77 Magdalena 2 Boiling 30 sec 71 4.54 Magdalena 3 Boiling 30 sec 68 4.66 67 Table B 1. Continued. Source Block Treatment Germination Percentage Standard Error Magdalena 4 Boiling 30 sec 83 3.76 Magdalena 1 Boiling 60 sec 63 4.83 Magdalena 2 Boiling 60 sec 68 4.66 Magdalena 3 Boiling 60 sec 63 4.83 Magdalena 4 Boiling 60 sec 82 3.84 Magdalena 1 Control 11 3.13 Magdalena 2 Control 6 2.37 Magdalena 3 Control 5 2.18 Magdalena 4 Control 4 1.96 Magdalena Percussion 1 min 32 4.66 Magdalena 2 Percussion 1 min 45 4.97 Magdalena 3 Percussion 1 min 46 4.98 Magdalena 4 Percussion I min 36 4.80 Magdalena 1 Percussion 2 min 67 4.70 Magdalena 2 Percussion 2 min 68 4.66 Magdalena 3 Percussion 2 min 65 4.77 Magdalena 4 Percussion 2 min 81 3.92 Magdalena 1 Percussion 4 min 87 3.36 Magdalena 2 Percussion 4 min 97 1.71 Magdalena 3 Percussion 4 min 83 3.76 Magdalena 4 Percussion 4 min 97 1.71 Magdalena 1 Percussion 8 min 98 lAO Magdalena 2 Percussion 8 min 100 0.00 Magdalena '3 Percussion 8 min 96 1.96 Magdalena 4 Percussion 8 min 98 lAO Manzano 1 Immersion 67 4.70 Manzano 2 Immersion 57 4.95 Manzano 3 Immersion 72 4049 Manzano 4 Immersion 73 4044 Manzano 1 Boiling 30 sec 70 4.58 Manzano 2 Boiling 30 sec 64 4.80 Manzano 3 Boiling 30 sec 71 4.54 Manzano 4 Boiling 30 sec 80 4.00 Manzano I Boiling 60 sec 50 5.00 Manzano 2 Boiling 60 sec 53 4.99 Manzano 3 Boiling 60 sec 59 4.92 Manzano 4 Boiling 60 sec 74 4.39 Manzano I Control 4 1.96 Manzano 2 Control 8 2.71 Manzano 3 Control 1 0.99 Manzano 4 Control 6 2.37 Manzano 1 Percussion 1 min 52 5.00 Manzano 2 Percussion 1 min 39 4.88 Manzano 3 Percussion 1 min 36 4.80 Manzano 4 Percussion 1 min 51 5.00 68 Table B 1. Continued. Source Block Treatment Gerrn.i~ation Percentage Standard Error Manzano Percussion 2 min 53 4.99 Manzano 2 Percussion 2 min . 58 4.94 Manzano 3 Percussion 2 min 61 4.88 Manzano 4 Percussion 2 min 62 4.85 Manzano 1 Percussion 4 min 76 4.27 Manzano 2 Percussion 4 min 85 3.57 Manzano 3 Percussion 4 min 79 4.07 Manzano 4 Percussion 4 min 83 3.76 Manzano 1 Percussion 8 min 87 3.36 Manzano 2 Percussion 8 min 83 3.76 Manzano 3 Percussion 8 min 83 3.76 Manzano 4 Percussion 8 min 75 4.33 Raton 1 Immersion 54 4.98 Raton 2 Immersion 37 4.83 Raton 3 Immersion 39 4.88 Raton 4 Immersion 69 4.62 Raton 1 Boiling 30 sec 36 4.80 Raton 2 Boiling 30 sec 20 4.00 Raton 3 Boiling 30 sec 28 4.49 Raton 4 Boiling 30 sec 60 4.90 Raton 1 Boiling 60 sec 15 3.57 Raton 2 Boiling 60 sec "'" 1.71 Raton 3 Boiling 60 sec 4 1.96 Raton 4 Boiling 60 sec 35 4.77 Raton 1 Control 4 1.96 Raton 2 Control 12 3.25 Raton 3 Control 2 1.40 Raton 4 Control 6 2.37 Raton 1 Percussion 1 min 52 5.00 Raton 2 Percussion I min 29 4.54 Raton 3 Percussion I min 35 4.77 Raton 4 Percussion 1 min 31 4.62 Raton 1 Percussion 2 min 55 4.97 Raton 2 Percussion 2 min 35 4.77 Raton 3 Percussion 2 min 33 4.70 Raton 4 Percussion 2 min 42 4.94 Raton 1 Percussion 4 min 65 4.77 Raton 2 Percussion 4 min 52 5.00 Raton "'" Percussion 4 min 51 5.00 Raton 4 Percussion 4 min 53 4.99 Raton 1 Percussion 8 min 88 3.25 Raton 2 Percussion 8 min 80 4.00 Raton 3 Percussion 8 min 73 4.44 Raton 4 Percussion 8 min 74 4.39 Sacramento 1 Immersion 43 4.95 69 - Table B 1. Continued. Source Block Treatment Germination Percentage Standard Error Sacramento 2 Immersion 30 4.58 Sacramento 3 Immersion 38 4.85 Sacramento 4 Immersion 56 4.96 Sacramento 1 Boiling 30 sec 47 4.99 Sacramento 2 Boiling 30 sec 48 5.00 Sacramento 3 Boiling 30 sec 48 5.00 Sacramento 4 Boiling 30 sec 56 4.96 Sacramento 1 Boiling 60 sec 52 5.00 Sacramento 2 Boiling 60 sec 67 4.70 Sacramento 3 Boiling 60 sec 43 4.95 Sacramento 4 Boiling 60 sec 48 5.00 Sacramento 1 Control 0 0.00 Sacramento 2 Control 3 1.71 Sacramento 3 Control 3 1.71 Sacramento 4 Control 0.99 Sacramento 1 Percussion 1 min 57 4.95 Sacramento 2" Percussion 1 min 61 4.88 Sacramento 3 Percussion 1 min 62 4.85 Sacramento 4 Percussion 1 min 55 4.97 Sacramento 1 Percussion 2 min 84 3.67 Sacramento 2 Percussion 2 min 70 4.58 Sacramento 3 Percussion 2 min 81 3.92 Sacramento 4 Percussion 2 min 69 4.62 Sacramento 1 Percussion 4 min 91 2.86 Sacramento 2 Percussion 4 min 84 3.67 Sacramento 3 Percussion 4 min 84 3.67 Sacramento 4 Percussion 4 min 94 2.37 Sacramento 1 Percussion 8 min 82 3.84 Sacramento 2 Percussion 8 min 81 3.92 Sacramento 3 Percussion 8 min 82 3.84 Sacramento 4 Percussion 8 min 81 3.92 Sandia 1 Immersion 76 4.27 Sandia 2 Immersion 70 4.58 Sandia 3 Immersion 70 4.58 Sandia 4 Immersion 81 3.92 Sandia 1 Boiling 30 sec 78 4.14 Sandia 2 Boiling 30 sec 71 4.54 Sandia 3 Boiling 30 sec 63 4.83 Sandia 4 Boiling 30 sec 78 4.14 Sandia 1 . Boiling 60 sec 67 4.70 Sandia 2 Boiling 60 sec 65 4.77 Sandia 3 Boiling 60 sec 74 4.39 Sandia 4 Boiling 60 sec 66 4.74 Sandia 1 Control 16 3.67 Sandia 2 Control 7 2.55 70 Table B 1. Continued. Source Block Treatment Gennination Percentage Standard Error Sandia 3 Control 8 2.71 Sandia 4 Control 16 3.67 Sandia 1 Percussion 1 min 67 4.70 Sandia 2 Percussion 1 min 61 4.88 Sandia 3 Percussion 1 min 66 4.74 Sandia 4 Percussion 1 min 59 4.92 Sandia I Percussion 2 min 82 3.84 Sandia 2 Percussion 2 min 83 3.76 Sandia 3 Percussion 2 min 81 3.92 Sandia 4 Percussion 2 min 84 3.67 Sandia 1 Percussion 4 min 89 3.13 Sandia 2 Percussion 4 min 86 3.47 Sandia 3 Percussion 4 min 89 3.13 Sandia 4 Percussion 4 min 85 3.57 Sandia 1 Percussion 8 min 80 4.00 Sandia 2 Percussion 8 min 77 4.21 Sandia 3 Percussion 8 min 69 4.62 Sandia 4 Percussion 8 min 66 4.74 LITERATURE CITED Ashby WC, Vogel WG, Rogers NF. 1985. Black locust in the reclamation equation. General Technical Report, Northeastern Forest Experiment Station, . USDA Forest Service. NE-I05, 12 p. Ballard LAT. 1973. Physical barriers to gennination. Seed Sci Technoll: 285-303. ___. 1976. Strophiolar water conduction in the seeds ofTrifolieae induced by action on the testa at nonstrophiolar sites. Austr J Plant Physiol3: 465-469: Barton LV. 1947. Special studies on seed coat impenneability. Contrib Boyce Thompson Inst 14: 355-362. Baskin CC, Baskin JM. 1998. Seeds: Ecology, biogeography and evolution of donnancy and gennination. Academic Press, London, UK, 666 p. Bewley JD. 1997. Seed germination and donnancy. Plant Ce119: 1055-1066. Bewley JD, Black M. 1994. Seeds: physiology of development and germination. 2nd ed. Plenum Press; New York, NY. 445 p. Briscoe CB. 1996. Growing Nitrogen Fixing Trees in Acid Soils. In: MH Powell (ed) Nitrogen Fixing Trees for Acid Soils- A Field ManuaL Winrock International, Morrilton, AR. 22 p. Copeland LO, McDonald MB. 1995. Principles of Seed Science and Technology. 3rded. Chapman & Hall: New. York, NY. 409 p. Cox, RE, Klett JE. 1984. Seed gennination requirements of native Colorado plants for use in the landscape. Plant Propag 30(2): 6-10. Dell B. 1980. Structure and function ofthe strophiolar plug in seeds ofAlbizia Lophantha. Am J Bot 67: 556-563. Dick-Peddie WA. 1993. New Mexico Vegetation: Past, Present and Future. University ofNew Mexico Press, Albuquerque, NM, 244 p. Dreesen DR. 2001. Propagation protocol for production ofcontainer Robinia neomexicana plants. Los Lunas Plant Materials Center, Los Lunas, New Mexico. In: Native Plant Network. URL: http://www.nativeplantnetwork.org (accessed 9 January 2002). Moscow (ID): University ofIdaho, CoIl Natural Resources, For Res Nursery. 72 Dreesen DR, Harrington JT. 1997. Propagation of native plants for restoration projects in the Southwestern US- Preliminary Investigations. In: Landis, TD; . Thompson, JR, tech coords. National Proceedings, Forest and Conservation Nursery Assocs. Gen Tech Rep PNW-GTR-419. Portland, OR: USDA, For Ser, PNW Res Sta: 77-88. Gottfried GJ. 1980. Control ofNew Mexican locust and the effect on planted ponderosa pine in central Arizona. US For Ser Res Note RM-386. Rocky Mountain For Range ExpStn, Fort Collins, CO. Gosling PG, Samuel YK, Jones SK. 1995. A systematic examination of germination temperature, chipping and water temperature/soak duration pretreatments on the seeds ofLeucaena leucocephala. Seed Sci & Technol 23:521-532. Hamly DH. 1932. Softening of the seeds ofMelilotus alba. Bot Gazette 93: 345 375. Hanna PJ. 1984. Anatomical features ofthe seed coat ofAcacia kempeana (Mueller) which relate to increased germination rate 'induced by heat treatment. New Phytol 96: 23-29. Hanks JP, Dick-Peddie WA. 1974. Vegetation patterns ofthe White Mountains, New Mexico. Southwest Naturalist 18(4): 371-382. Hine S Harrington JT, Dreesen DR. 1997. Seed scarification requirements for Robinia neomexicana. In: Harrington, JT,tech coord. Proc Third SouthwestemContainer Growers Conf. Feb 12-13, 1997. Ft. Collins, CO. Hopkinson JM. 1993. The strophiole of leguminous seeds. Newsletter- IntI Herbage Seed Production Res Grp 18: 7-8 . ----. 1997. Differences in hardseededness between Cunningham and Tarramba leucaena. In: Leucnet News: The Newsletter of the International LeucaenaRes and Devlt Netwk. No.4. ISTA (International Seed Testing Association). 1999. International rules for seed testing rules. Seed Sci & Technol27: Supplement. Kelly KM, van Staden J. 1987. The lens as the site of permeability in the papilionoid seed, Aspalathus linearis. J Plant Physiol 128(4/5): 395-404. 73 Keresztesi B. 1988. Black locust: the tree of agriculture. Outlook on Agr 17(2): 77~85. Kigel 1 and Galili G. 1995. Seed development and germination. Marcel Dekker, New York, 853 p. Klemmedson 10. 1994. New Mexico locust and parent material: Influence on macronutrients of forest floor and soil. Soil Sci Soc Am 1 58: 974~980. Kuo WH. Tam AW. 1988. The pathway ofwater absorption ofmungbean seeds. Seed Sci & Technol16(1): 139~144 .. Leadem CL. 1997. Dormancy--unlocking seed secrets. In: Landis TD. Thompson lR, tech coords. National Proceedings, Forest,and Conservation Nursery Assocs. Gen Tech Rep PNW-G TR-419. Portland, OR: USDA, For Ser, PNW Res Sta: 43-52. . Lin Y, Wagner MR, Seweyestewa A, Honahni MD, DeGomez M" McMillin JD. 1996. Influence ofcultural treatments on seed germination and seedling growth ofRobinia neomexicana. Forest, Farm and Community Tree Res Reports, Volume 1. Marunda CT. 1990. Effects ofseed pretreatment on the development ofAcacia auriculforinis and A. holosericea seedlings. ACIAR Proceedings Series No. 28. Meginnis HG. 1937. Sulphuric acid treatment to increase germination ofblack locust seed. USDA Circ. 453. 34 p. Miklas PN, Towensend CE, Ladd SL. 1987. Seed,coat anatomy and the scarification ofcicer milkvetch seed. Crop Sci 27: 766-772. Monsen SB. 1984. Use ofshrubs on mine spoils. In The Challenge ofProducing Native Plants for the Intermountain Area: Proceedings: Intermountain Nurseryman's Association 1983 conference, Las Vegas, NV, compo PM Murphy, 26-31, Ogden, UT: USDA For Ser Gen Tech Report INT-l68. Olson DF. 1974. Robinia L. Locust. In: Seeds of Woody Plants in the US, tech coord. CS Schopmeyer, 728-731. USDA For Ser Agr Handbk No.450. Rehman S, Loescher RNl, Harris P1C. 1998. Dormancy breaking and germination ofAcacia salicina Lindl. seeds. Seed Sci & Techno127: 553-557. Rolston MP. 1978. Water impermeable seed dormancy. Bot Rev 44: 365-396. 74 SAS Institute Inc. 1989. SAS/STAT User's Guide, Version 6, 4th Edition, Volume 1. Cary, NC: SAS Institute Inc. 943 p. Serrato ..Valenti G, Conara L, Ferrando M, Modenesi, P. 1993. Structural and histochemical features ofStylosanthes scabra (Leguminosae; Papilionoidae) seed coat as related to water entry. Can J Bot 71: 834-840. Simpson BJ. 1988. A field guide to Texas trees. Austin, TX: Texas Monthly Press. 372p. Tran VN, Cavanagh AK. 1984. Structural aspects ofdormancy. In: "Seed Physiology" Vol. 2, Murray, DR (ed), p.1-44. Academic Press, Australia. USDA Forest Service. 1948. Woody-plant seed manual. USDA Misc Pub1654, 416p. Vines RA. 1960. Trees, Shrubs and Woody Vines ofthe Southwest. Austin: Univ ofTexas Press. 1104 p. Wagner MR, Larson RE, Craig TP. 1992. Ecology and silvics ofNew Mexico locust (Robinia neomexicana) with an emphasis on the arboreal form, p.53-57. In: JW Hanover, K Miller, and S Plesko (eds), Proc, Intemat Confon Black Locust: Biology, Culture and Utilization. Mich State Univ, USA. Wilson JK. 1944. Immersing seeds ofspecies of Robinia in boiling water hastens germination. J For 42: 453-454. Zimmerman RW, Carpenter SB. 1980. First-year coppice production from a 5 year old black locust stand on surface mine spoil, p.309-314. Centr Hardwd For ConfProc. University ofMissouri, Columbia.