POSTER SESSION

L. Castro-Acero1, J.J. Marquez-Ortiz2, and J. Santamaria-Cesar2
National Inst. of Forestry, Agriculture and Livestock Research (INIFAP)
1 C.E. Valle de Mexico, Apdo. Postal 10, CP 56230 Chapingo, Mex., Mexico.
2 C.E La Laguna, Apdo. Postal 247, CP 27000 Torreon, Coah., Mexico.

Alfalfa (Medicago sativa L.) was introduced in Mexico from Spain in the 16th century. Four hundred years of natural selection resulted in Mexican alfalfa ecotypes. Cultivars selected from Mexican ecotypes have good local adaptation and higher winter yields than cultivars from California or Spain (Castro-Acero, 1978). In the 1970s the National Institute for Forestry, Agriculture and Livestock Research (INIFAP) conducted trials of Mexican germplasm collections to asses their productivity. Data from these trials are a valuable basis for designing further studies involving collection and characterization of alfalfa germplasm in Mexico. The objective of this research was to characterize and classify Mexican alfalfa germplasm based on agronomic traits.
The experiment was established in January 1976 at the Valle de Mexico Experiment Station in Chapingo, Mex. The trial included 46 ecotypes from Atlixco, Pue.; Atoyac, Jal.; Navojoa, Son.; Oaxaca, Oax.; San Miguel Octopan, Gto.; Tanhuato, Mich; and Zacapu, Mich. Control cultivars were Moapa and San Joaquin II (California) and Aragon (Spain). Accesions were evaluated under a 7 x 7 lattice design with four replications. The experimental unit consisted of three 5-m rows separated at 30 cm. Twenty four harvests were taken during a three-year period. We measured dry matter yield; symptoms of leaf spot, downy mildew, virus, spotted aphid, and pea aphid; plant height; uniformity; leafiness; regrowth height; plant vigor; and plant survival. Variance and correlation analyses were used to select classification variables. Variable standarization, principal component analysis, and average linkage cluster analysis were used to determine groups based on dendrogram interpretation.
Ecotypes had significant effects (P<0.05) for all variables. We found six different clusters (number of ecotypes in parenthesis): Zacapu+Atlixco (15), Zacapu+San Miguel (6), Oaxaca-1 (11), Tanhuato (13), Oaxaca-2 (1), and Navojoa (3). Clusters Oaxaca-1 and -2, Tanhuato and Navojoa coincided well with their area of collection. The cluster Tanhuato had high yield, low virus and aphid damage, and highest plant height, regrowth, vigor and plant survival scores. Moapa and San Joaquin II fell within Oaxaca-1, a cluster with low adaptation, suggesting that California cultivars were as low in adaptation as ecotypes from Oaxaca. Cultivar Aragon was similar to ecotypes from an array of locations in central Mexico, indicating that such ecotypes are still similar to those from Spain. This research suggests that local ecotypes have potential to enhance agronomic performance of alfalfa in Mexico.
References

Castro-Acero, L. 1978. Alfalfa. p. 165 In T. Cervantes S. (ed.) Recursos
geneticos disponibles a Mexico. SOMEFI. Chapingo, Mex.

Alfalfa On-Farm Research at North-Central Mexico

J.J. Marquez O., G. Nunez H., Y.I. Chew M., E. Castro M. and M. Ramirez D.
National Inst. of Forestry, Agriculture and Livestock Research (INIFAP)
C.E La Laguna, Apdo. Postal 247, CP 27000 Torreon, Coah., Mexico.

Alfalfa is the main forage for dairy farms in the temperate and arid zones of Mexico, where 290 000 ha are grown under irrigation. La Laguna has the most advanced dairy industry in North-Central Mexico, but, due to water availability problems, the amount of alfalfa grown in this region has never surpassed 30 000 ha. Recently, local farmers created La Laguna Foundation for Agricultural and Livestock Research (PIAL). PIAL coordinates and supports private (LALA, Inc.) and public (INIFAP) research efforts to increase forage crop productivity. Integrated Alfalfa Management, a current project supported by PIAL, involves an on-farm component to promote and increase technology adoption and design research according to farmer s needs. Our objective is to develop an integrated management system to increase alfalfa productivity and persistence under our local environmental conditions.
In 1995, our research involved introduction of alfalfa nondormant cultivars; diagnose and control of diseases, insects, and weeds; and postharvest management. Ten cultivars were evaluated (1 x 3 m plots) in replicated field trials at three locations. Twenty farms were monitored for disease incidence in March, June, August, and November. Ten farms were monitored for beneficial and damaging insects. Six preemergent and seven postemergent herbicide treatments were tested in two separate trials at one location to determine their ability to control dodder ( Cuscuta sp). Alfalfa, as hay or silage, with or without monensin, was fed to cows to assess treatment effect on dry matter intake during the hot summer season.
Cultivars had a significant (P<0.05) effect on first year fresh and dry matter yield of alfalfa. The highest yielding cultivar was SW-14; the lowest, Nitro and Aragon. Diseases were crown rot (Fusarium, Rhizoctonia, Phoma, Colletotrichum), anthracnose (Colletotrichum), and root rot (Phymatotrichum omnivorum). Crown rot was significantly (P<0.05) correlated with stand age. Nematodes (Pratylenchus, Helicotylenchus, Tylenchorynchus, Xiphinema) were not a serious problem. Damaging insects found included aphids (Therioaphis, Acyrthosiphon), leafhoppers (Empoasca), Diabrotica, thrips (Frankliniella) and false bug (Lygus). Beneficial insects were minute pirate bug (Orius), ladybug (Hippodamia), bigeyed bug Geocoris and Braconidae. Preemergent herbicides TR-10 (10-20 kg/ha) and Prowl-400 (3-6 l/ha) controlled dodder efficiently; postemergent herbicides did not. Forage losses were smaller for silage (13.4%) than for hay (33.9%). Dry matter intake and milk production and quality did not differ when cows were fed either silage or hay with or without monensin.
New research involves agronomic and nutritive characterization of germplasm available in Mexico, estimation of alfalfa losses caused by crown rot, and screening germplasm for local diseases.

The Yield and Ouality of the U.S.A. Multileaf Alfalfa Varieties In Poland

G. Harasimowicz-Hermann*, J. Andrzejewska*, W. Nowak**, F. Gospodarczyk** and W. Waniorek***

*University of Technology and Agriculture, Mazowiecka St., 8S-084 Bydgoszcz
**University of Agriculture, Norwid St., 50-375 Wroclaw
***Agricultural Experiment Station, 64-312 Brody, Poland

The objective of the study was to determine the usefulness of U.S.A. multileaf alfalfa varieties for cultivation under varied Polish soil and climatic conditions. Multileaf forms differ from the others because up to 30% their leaves consist of 4-7 leaflets instead of a typical trifolium. Alfalfa variety 'Legend' and the genera FG3B50 and FG4B59, breeded by Dr. Mark McCaslin, Forage Genetics, West Salem, WI, were tested. Two series of three-year-long multipoint field experiments were settled in 1993 and 1994. The U.S.A. varieties were compared with the standard Polish variety 'Radius', well prepared for Polish conditions. Alfalfa yields of green and dry matter were assayed and evaluated statistically. Total protein, fiber and soluble sugars content, digestibility in vitro, winterhardiness, regrowth rate, as well as stand density were also assessed. Determinations made for I series (1993-95) include the seeding ycar, and two production years, while those made for II series (1994-95) the seeding year and first production year. The highest yields of green and dry matter and protein were found for I series at Brody where alfalfa varieties were seeded alone. The experiment was conducted on brown soil developed from boulder clay. Average rainfall for that period was 765 mm and the temperature 8.5 deg C. High yields were also noted near Wroclaw, where alfalfa was cultivated on brown soil of typical grey-brown podzolic type. In the period of study average rainfall was 1750mm and the temperature 8.5 deg C. Most unfavourable soil and climatic conditions occur at the experiment site ncar Bydgoszcz, where the lowest alfalfa yields were observed. The soil type is pseudo podzolic on strong loamy sand. The average rainfall was 443 mm and the temperature 8.3 deg C. The results of I series indicated that the level of yields and forage quality of multileaf 'Legend' were close to those of the standard 'Radius', while 'Legend' was significantly better in II series. In the seeding year higher yields, especially in case of green matter, were noted for 'Legend'. Very good soil, moisture and thermal conditions at Brody promoted high and equalized yields of all the varieties studied. In production years the climate of south-west Poland caused better performance of the U.S.A. varieties, especially 'Legend'. Under most difficult soil and climatic conditions of central Poland 'Radius' yielded higher in second production year than the U.S.A. varieties. Total protein yields (I series) for 'Radius' and 'Legend' were 4345 and 4370 kg/ha, respectively, while the corresponding sequence for II series was: 'Legend'>FG3B50>Radius>FG4B59 (2253, 2043, 1997, 1994 kg/ha, respectively). Fodder quality of all varieties was similar, with one exception for FG4B59, for which- the highest protein yield was found. Changing soil and climatic conditions did not affect the fodder quality. Winterhardiness and regrowth rate ofthe U.S.A. varieties did not vary from the Polish one.

The Improvement of alfalfa Breeding in China

Zhensheng Gao, Xueju Xie, Qidong Ma and Fuzeng Hong
Grassland Institute, China Agricultural University, Beijing 100094, China

Alfalfa is the most widespread cultivated and applied legumes around the world which had the highest utilization value. China has more than 2,000 years history in introduction of alfalfa, but breeding and selection of new varieties is a relatively new event in China. The selection and introduction of important herbage cultivars was started in the fifties in order to advocate the rotation of grassland crops and to develop the animal husbandry, such as 'Gongnong No.l' and 'Gongnong No.2' Medicago sativa L., which was bred by single and mixed selection method, based on the introduction of 'Grimm' alfalfa at the end of fifties. In the seventies, researchers used local M. sativa L. cross wild M. falcata to enhance the cold tolerance and drought resistance, bred 'Caoyuan No. I' and 'Caoyuan No.2' M. varia Martin. with the started working on Chinese herbage cultivar licensing and registration since 1986, researchers has compared and evaluated the alfalfa around China, which had appropriated to grow in the local areas after long time of adaptation, collection and collation a series of local alfalfa varieties.

There are twelve species and three formas in the genus, which are grown in China. The most commonly grown species are Medicago sativa L., M. falcata L., M. media Pers, M. ruthenica L., M. hispida Gaerth. and M. lupulina L. M. sativa and M. hispida is common]y grown in the south of China. Nowadays, China has totally 30 licensed alfalfa cultivars, including 23 M. sativa L. cultivars and 7 M. varia L. cultivars.

Chinese Herbage Cultivar Registration Board (CHCRB) was established in 1987. From 1986 to 1994, CHCRB has registered totally 163 herbage, fodder crops and turfgrasses, including 78 grasses, 70 legumes and 15 other families. The proportion of alfalfa cultivars in total registered cultivars is 18.4% and 38.5% in total legumes cultivars, including 11 newly bred cultivars, 17 local cultivars, 1 wild species and I introduced cultivar. Most ofthe 17 local cultivars are M. sativa L. There are 1.3 million hectares alfalfa plant areas in Chilla, mainly distributed in Northwest 66.8%, North China 30.3% and Northeast 2.7%, licensed cultivars in Northwest is 40%, 30% in North China and 23% in Northeast of China.

References
1. Edited by the Chinese Herbage Cultivar Registration Board (CHCRB), 1992, LICENSED CULTIVARS OF HERBAGE CROPS IN CHINA, Beijing Agricultural University Press, Beijing, China.
2. Edited by CHCRB, 1995, Working report of Chinese Herbage cultivar Registration Board.

The Study on Isozymic of Local Alfalfa Cultivars in China

Xueju Xie, Mojun Chell and Zhenshellg Gao
Grassland Institute, China Agricultural University, Beijing 100094, China

Alfalfa were introduced into China about two thousands years ago. It was widely spread through out provinces and regions of north of the Changjiang River. The sources of local alfalfa cultivars are very rich in China. Thirty alfalfa cultivars were licensed by the Chinese Herbage Registration Board at the end of 1995, including 17 local cultivars which belong to the M. sativa L.. In the palmer, we collected total 17 local cultivars and domesticated wild cultivar (M. media Pers.). Esterase isozyme of single plant of every cultivar was analyzed, in order to find out the esterase isozyme band -l types, and the relationship between different cultivars. th result showed: total 14 isozyme bands can be obtained in four leaves stage. Among them, 7A, 12A, 13A, and 14A are the public bands of alfalfa. But the emerge frequency of other bands is different in every cultivar. Correlation analysis on 11 bands and 7 meteorological factors showed that the frequency of 2A, 3A, 4A, 5A and 8A had correlation with average year temperature, average lowest temperature, average highest temperature and >5 °C year temperature summation.

Statistical analysis of esterase isozyme of 900 materials from single plant of every cultivar showed that Est-zymogram types have 60 patterns, 32 patterns are specific types of cultivars. Est-zymogram types are not only varied, but also abundant among cultivars and different plants of every cultivar. Experimental evidences indicate that the gene resources of alfalfa are very rich in China.

Cluster analysis on esterase isozyme of 18 cultivars showed that relationship was close among Pianguan, Hexi, Aletai, Beijing and Cangzllou, so did that among Longdong, Guangzhou, wudi, Yuxian, Longzhong, Ohan, Huaiyin, Tianshui, Neimengzhunger and Xinjiangdaye are more and more distance successively. Neimengzhunger and Xinjiangdaye are the most distant from others.

References

Gao Zhenslleng et al .1994, Study on the comparison of Esterase Isozyme between Creeping rooted and Non-creeping -rooted Alfalfa. Report of the thirty-fourth North American Alfalfa Improvement Conference. Guelph . Canada . 142.

GENETIC DIVERSITY IN WILD ACCESSIONS OF MEDICAGO SATIVA FROM SPAIN.

E. Jenczewski, G. Genier, J. Ronfort and l.M. Prosperi.

INRA- Station de Genetique el Amelioration des Plantes, F-34130 Mauguio. France

Rearing is a traditional component of the Mediterranean agriculture system. Nowadays, it also represents the best way to valorize population-drained marginal areas for agricultural, landscape maintenance and spare-time purposes. Improvement programs thus require to focus on hardy and grazing-adapted fodder types; pre requisites are firstly the collection and evaluation of wild ecotypes diversity.
One hundred and four natural populations of Medicago sativa (called "Mielgas") were collected in Spain, from September 1985 to July 1987, with the collaboration of I. Delgado Enguita (SIA-DGA, Zaragoza, Spain). They were mainly from roadsides, unirrigated and often grazed rangelands, or even orchards (olive trees, vineyards...).
This germplasm has been evaluated at Montpellier, since 1986, under both grazing and cutting managements. Most of the <> plants have a prostrated habit and .l great ability of soil colonization (rhizomes). Assessing the agronomic variability of the sampled populations notably prove that this germplasm displays an important variability in growth rhythm (the first cut in second year represents from 39% to 75% of the total annual dry matter production, which itself accounts for 60% of the standard cultivars production), a greater perenniality than alfalfa cultivars and a good vegetative spring yield and seed production level for non-selected populations. Multivariable analysis (figure 1) also demonstrate that the populations from the North-East of Spain can be easily distinguished from the other "Mielgas", since they behave as cultivated types.

Genetic diversity was also implemented through the survey of sixteen populations with thirty RAPD markers. These populations were chosen in order to be representative of the diversity which arose from the former analysis. The distribution of RAPD polymorphism was assessed by partitioning this diversity within and among populations components. This study confirmed the remarkable intra-population diversity of <<Mielgas>>; indeed, in a previous study concerning two populations (wild population/spanish cultivar), thirteen primers uncovered sixty five fully reproducible fragments. Most of them were polymorphic (93%) and indicated that 95 per cent of the variation they displayed was partitionated within each population.
References

Delgado Enguita I., 1989. Estudio de la Variabilidad de las Mielgas Aragonesas en areas de precipitation anual inferior a 600 mm. Tesis Doctoral-Universidad Polytechnica de Madrid, Spain

Prosperi J. M., Delgado Enguita I., Angevain M., 1989: Prospection du genre Medicago en Espagne et au Portugal. Plant Genetic Resources Newsletter 78/79, 27-29.

Prosperi J.M., Angevain M., Mansat P., 1990: Valorization of forage genetic resources for selection of adapted cultivars. Example of wild mediterranean lucernes. 95-98. in 6th Meeting of the FAO European sub-network on Mediterranean Pastures and Fodder Crops. October 17/19, 1990. Bari (Italy).

Effect of Grazinq Frequency and Intensity on Yield and
Persistence of Alfalfa-Tall Fescue Pastures.

Nestor A. ROMERO and N. Tony JUAN
Anguil Experiment Station, INTA
La Pampa (6326), Argentina.

There are five millon ha of alfalfa in the Pampean Region of Argentina. In about 75 % of this area alfalfa is in mixture with perennial grasses, mostly tall fescue (Festuca arundinacea L.). The use of moderately dormant to nondormant alfalfa cultivars has increased since the early 80s. These cultivars regrow faster after cutting or grazing, have higher forage production in early spring and late fall, reach maturity faster, and have different plant structure than dormant alfalfas. Differences among alfalfa genotypes could affect alfalfa-grass competition and generate differential responses to a given planting and grazing system. The objective of this study was to evaluate the effect of alfalfa fall dormancy, row distance, and grazing system on yield and persistence of alfalfa-tall fescue mixtures.
Twelve treatments were evaluated in a 2x3x2 factorial experiment, factors were arranged in split blocks with six 1.6-ha paddocks as blocks. Factors were: 1) Grazing system (GR1, 3,5 days of grazing + 17,5 days of resting; stocking rate adjusted to graze tall fescue to half its height; and GR2, 7 days of grazing + 35 days of resting; stocking rate adjusted to leave a 5-cm pasture residue); 2) Alfalfa fall dormancy (NDOR, nondormant cultivar WL 605, FGS=9; MDOR, moderately dormant I cultivar WL 320, FGS=4; and DOR, dormant II ecotype Pampeana, FGS=3); 3) Row distance (ROW15, 15 cm between alfalfa and tall fescue alternate rows; and ROW30, 30 cm). Forage yield (Mg DM ha-l) was estimated at the beginning of each grazing period during three growing seasons. Persistence (ground cover as a % of initial cover) was measured once a year.
Alfalfa yield was increased and tall fescue yield was decreased under the GR2 grazing system as compared with GR1. With frequent but not intense grazing (GR1) both alfalfa and fescue yielded more in ROW15 than in ROW30. With less frequent but more intense grazing (GR2), alfalfa also yielded more in ROW15, but the opposite happened with tall fescue. Yield of DOR alfalfa was lower than yield of MDOR and NDOR alfalfa with both row distances. Within ROW15, tall fescue yielded more associated with dormant alfalfa than with less dormant ones. Persistence of tall fescue was higher under GR1 than under GR2, regardless of alfalfa fall dormancy. All alfalfa genotypes had higher persistence under GR2 than under GR1. Persistence of DOR alfalfa was always lower than that of MDOR and NDOR alfalfa.
We conclude that, for our conditions, total yield of alfalfa-fescue mixtures was affected by alfalfa dormancy and row distance, but not by grazing system. Yield and persistence of alfalfa was increased when a moderately dormant to nondormant genotype was planted at 15 cm between rows and managed with a low frequency/high intensity grazing system. Yield and persistence of tall fescue was highest when it was planted with a dormant alfalfa at 30 cm between rows, and grazed frequently but not intensively.

INBREEDING AS A METHOD OF DEVELOPING HOMOZYGOUS LINES IN
ALFALFA, MEDICAGO SATIVA

E. V. Kvasova
Institute of Cytology and Genetics, Siberian Branch of the Russian Academy of Sciences,
630090, Novosibirsk, 90, Russia

Applied genetics of self-incompatible species, such as alfalfa, has been insufficiently developed because of the complex genetics nature of their populations. It is known that some problems of applied genetics and breeding can be solved by using homozygous lines. With this in mind we produce homozygous lines by the method of inbreeding.

The results obtained with artificial population showed that alfalfa inbred lines can be obtained only when a great number of plants from populations are grown and analyzed. Thus, the total number of alfalfa depressed plants in the first inbred generation was 83.4%. Plant vigor started to recover after the third inbred generation. All the characters differentiate in the first 2-3 generations. The differences in morphological characters were wider among plants within than between lines. The widest phenotypical variations in shape and size of leaf, flower and raceme, stem structure were noted in the third generation. There were also variations in the size and shape of pollen grains. The observed differences reduced with each successive inbred generation so that the plants within the lines became homogeneous by the 7-8th generations and the differences between lines became clear-cut.

Based on the obtained results, it is concluded that inbreeding allows to identify forms with any desirable character in a population. In development of inbred lines, the aim of the experiment must be clearly defined for the start. A group of self-fertile plants in a population suffices when self-fertile forms are to be obtained. The greatest possible part of the population gene pool must be involved when more diverse inbred lines are to be produced. This can be achieved by a strategy overcoming the effect of self-incompatibility system. This is because the group of selling plants from which the inbred lines will be derived is too small for involving the population gene pool.

WEEDALF$: An Integrated Expert System for Weed Management in Alfalfa

J. F. Stritzke, C. W. Cuperus, and S. Zhang
Department of Agronomy and Entomology, Oklahoma State University
Stillwater, Oklahoma, 74078, U.S.A.

WEEDALF, the first version developed in 1989 dealt with the cost and benefits of using herbicides in established alfalfa. The new version, WEEDALF$ (Version 2.0) has been upgraded to include an integrated approach to weed management in both seedling and established stands of alfalfa. The four main tasks of the system are weed management, weed information, herbicide information, and profitability of using herbicides. Each component can be accessed individually, but all are interconnected and can be utilized in other sections.
With the "weed management" section, there is much accumulation and integration of information to determine the weed problems and then there is a listing of management alternatives. For example, with seedling stands planted in the fall, integrated knowledge involves previous crop, weed problems in previous crop, weed control used in previous crop, type of seeded preparation, soil pH and fertility level, and type and extent of weed problems. When weed control by herbicides is advised, the user has the option of selecting from a listing of available labeled herbicides. At this point in the program, the system switches to the herbicide information section and allows the user to view general description and comments on the various herbicides. Selection of the various herbicide options will result in a cost/benefit display and an estimated profit with each option.
The "weed information" section contains knowledge about weeds such as type, characteristics, and control options. The "herbicide information" section contains description of herbicides currently labeled on alfalfa. Information on each herbicide includes common name, trade name, labeled use rate, current suggested retail cost, alfalfa growth stage approved for application, growth stage of weed when effective, and some general comments dealing with its application and effectiveness. The "profitability of using herbicide" section is designed to calculate profitability of a herbicide treatment compared to an untreated check.
The software was developed as a part of Oklahoma State's ongoing Integrated pest Management program. The system is designed by using the PAD-BASED model based on prototypes and delegation mechanism concepts, and developed in the KnowledgePro(R) environment. The system requires a minimum of 640K of RAM with DOS 3.0 or later version. A color monitor is recommended but not required.

Breeding a Bloat-Tolerant Alfalfa in Argentina

D. H. Basigalup, C. V. Castell and C. D. Giaveno
EEA Manfredi/EEA Anguil - I.N.T.A. - Argentina
Partially financed by PRODUSEM, Pergamino, Argentina

With nearly 6.5 million ha, Argentina is the second largest alfalfa producer in the world. Approximately 90% of this area is devoted to beef and dairy production under direct grazing; therefore, bloat becomes a very serious problem every year. An economical solution would be to have bloat-tolerant alfalfa cultivars adapted to Argentine conditions. In 1991, the National Institute for Agricultural Technology (INTA) initiated a breeding program to develop such cultivars. In this paper we will briefly present that program and discuss the progress obtained up to now.

The breeding goal is to select non-dormant cultivars with a significant reduction in bloat potential. The objectives are: 1. To reduce the initial rate of ruminal disappearance (IRRD); and 2. To maintain a good forage quality level in the lower IRRD population.
Using the modified nylon bag technique (1), the program combines phenotypic and genotypic selection (polycross progeny test) for reduced IRRD after 4 h in the rumen of fistulated steers. In 1995, seed of the lst cycle of selection (C-l) was produced in pollination cages. Data from the progeny test were also used to estimate IRRD broad sense heritability (h ) based on variance component analysis.
Approximately 1300 C-l plants wer& transplanted to the field in September 1995 to initiate the 2n cycle of selection. Other 60 C-l plants were contiguously transplanted to 60 C-0 plants in order to estimate selection progress on IRRD reduction and to compare forage quality based on FRRD (= final rate of ruminal disappearance after 72 h in the rumen), L:S (= leaf:stem ratio), CP (= crude protein), ADF (= acid detergent fiber), NDF (= neutral detergent fiber); and RFV (= relative feed value). The expected response to selection (R) was estimated based on the equation: R = h . S, where h i s the heritability for IRRD and selection differential.
Calculated mean h2 was 0.18 and S was 8.74%; therefore, R was estimated to be 1.58%. The realized response to selection, based on the comparisons between C-l and C-0 populations, was a reduction of 1.98% in IRRD. C-l plants exhibited a significant higher fiber content than C-0 plants, but this increase is still away from any detrimental effect on forage quality. There was as much variability for all traits in the C-l population as in the C-0 population, indicating possibilities for improving any trait if necessary.
Reference

1. HOWARTH, R.E., B.P. GOPLEN, S.A. BRANDT, and K.-J. CHENG. 1982. Disruption of leaf tissues by rumen microorganisms: An approach to breeding bloat-safe forage legumes. Crop Sci. 22:564 568.&1

Genetic Variation for Digestibility and Fiber Contents in the Medicago sativa Complex

B. Julier and C. Huyghe
INRA, Station d'Amelioration des Plantes Fourrageres, 86600 Lusignan, France

Alfalfa energetic value is limited by its low digestibility. Genetic progress is necessary to increase or maintain the use of alfalfa forage in ruminant feeding under intensive conditions. Breeding for quality implies genetic variation for digestibility and related traits such as fiber contents, and a breeding criterion. Improvement of forage digestibility must be done without decrease in forage yield potential.
Digestibility and fiber contents were analyzed on 25 alfalfa populations of the Medicago sativa complex. Populations from ssp. sativa and ssp. falcata, diploid or tetraploid, were used. These populations proved to cover a large range of variation for morphological traits (2). They were sown at Lusignan (France) in a 2-rows plot design with 4 blocks in 1993. First cuttings in spring in 1994 and 1995 were harvested. Forage yield was determined, and stems were separated in 1994. Samples (stems in 1994, forage in 1995) were dried, ground and analyzed for enzymatic digestibility and fiber contents (NDF, ADF and ADL). Digestibility of NDF and lignin content in the cell wall were calculated.
Analysis of variance were performed on the traits, and coefficients of correlation calculated to determine tlle traits useful in breeding, i.e. traits that were correlated with digestibility but not correlated with forage yield. Genetic variation for digestibility was assessed? taking into account the forage yield. Analysis of variance was performed on the residues of the regression between digestibility and forage yield.
Wide genetic variation exits for these quality traits, at tlle sub-species level and at the population level. Fiber contents (NDF, ADF and ADL) were closely linked to digestibility but digestibility was negatively correlated with forage yield. Digestibility of the cell wall and cell wall lignin were correlated with digestibility and independent of forage yield. They could be valuable breeding criteria to improve alfalfa digestibility without reduction in forage yield. On average, at a given level of forage yield, populations of the falcata sub-species were more digestible and had lower fiber contents than populations of the sativa sub-species. Improvement of cultivated alfalfa could involve crossings witl1 tl1e wild subspecies falcata. These results confirm those already published (3). However, other authors (1, 4) observed lower digestibilities in ssp. falcata related populations than in ssp. sativa populations.
References

1. Buxton D.R., Hornstein J.S., Martin- G.C. (1987). Genetic variation for forage quality of alfalfa stems. Can. J. Plant. Sci. 67, 1057-1067.
2. Julier B., Porcheron A., Ecalle C., Guy P. (l995). Genetic variability for morphology, growth and forage yield
among perennial diploid and tetraploid lucerne populations (Medicago sativa L.). Agronomie 15, 295-304.
3. Julier B., Guy P., Castillo-Acuna C., Caubel G., Ecalle C., Esquibet M., Furstoss V., Huyghe C., Lavaud C.,
Porcheron A., Pracros r., Raynal G. (1996). Genetic vvariation for disease and nematode resistance and forage quality
in perennial diploid and tetraploid lucerne populations (Medicago sativa L.). Euphytica, in press.
4. Lenssen A. W., Sorensen E. L., Rosler G. L., Herbers L. H. (1991). Basic alfalfa germplasms differ in nutritive content of forage. Crop Sci. 31, 293-296.

Biotechnological Improvement of Alfalfa Nutritive Quality by Anti-sense Expression of
COMT and CCOMT. Methylating Enzymes in Lignin Biosynthesis

V. J. H. Sewalt, J. W. Blount, and R. A. Dixon

Lignin, a plant cell wall polymer derived from phenylpropanoid precursors with varying numbers of aromatic methoxyl groups, interferes with forage digestion to an extent that is dependent on its monomer composition (1). Our objective is to improve alfalfa nutritive quality by reducing lignin concentration and manipulation of extent of lignin methoxylation. We are also working towards elucidation of the relative importance of different methylating enzymes (O-methyltransferases) in monolignol biosynthesis.
Experiments were inititiated to manipulate lignin in alfalfa by downregulation of COMT (caffeic acid O methyltransferase) and CCOMT (caffeoyl-CoA O-methyltransferase) using anti-sense technology. A number of single and double anti-sense constructs with COMT and/or CCOMT under control of the dual 35S promoter were introduced into alfalfa by Agrobacterium-mediated transformation. Controls were untransformed, but regenerated plants, and plants harboring 35S-GUS (pBI121 plasmid). The two controls were not different in OMT activities or lignin characteristics.

A number of plants were obtained in which COMT, CCOMT, or both OMTs were reduced. Most antisense plants with reduced OMT activity and all plants harboring pBI121 were MPT II-positive by ELISA-assay. Down-regulation of COMT and/or CCOMT in several primary transformants resulted in modestly reduced lignin concentration and/or extent of methoxylation. One single COMT antisense transformant was strongly reduced in lignin, and its lignin displayed a sharp increase in methoxyl content, indicative of an increase in S/G ratio. Interestingly, cell wall (NDF) concentration was drastically reduced in many lines. Enzymatically determined digestibility of NDF improved only slightly in many lines, but due to the reduction in NDF content, most transgenic lines had significantly higher dry matter digestibility. Further lignin analysis of promising lines and confirmation of transgene expression and copy number is currently underway. The preliminary results of this experiment demonstrate the feasibility to generate improved alfalfa lines with reduced lignin and altered lignin composition for improvement of digestibility.
Reference

1. Sewalt, V. J. H., W. G. Glasser, J. P. Fontenot, and V. G. Allen. 1996. Lignin impact on fiber degradation. 1.
Quinone methide intermediates formed from lignin during in vitro fermentation of corn stover J. Sci. Food
Agric. (in press).

ALFALFA YIELD RESPONSE TO METHOD AND RATE OF APPLIED P AND
WEED INTENSITY

S.B. Phillips, G.V. Johnson, and W.R. Raun
Oklahoma State University, Stillwater, Oklahoma

ABSTRACT

Past experience in Oklahoma indicates that initial and maintained soil fertility levels strongly affect long-term production of alfalfa (Medicago sativa L.). Presently, there are indications which suggest that P fertilizers should be applied at two to three times the recommended rate at the time of establishment when the fertilizer can be incorporated. Preplant band applied P fertilizer may provide increased long term benefit if soil-fertilizer P reactions are reduced. High rates should provide a P fertility foundation with the potential for sustaining yields for several years. Other work suggests that carrier N in some sources of P would promote weeds if high P rates were used. The objectives of this work were to evaluate the effect of alternative methods of applying phosphorus fertilizer on alfalfa yield and weed populations and to determine optimum phosphorus fertilizer rates and timing of application for alfalfa production. The experiment was established at the South Central Research Station in Chickasha, OK, on September 9, 1992. Since the trial was initiated, five forage harvests have been obtained each year beginning May 5, 1993. A linear response to P has been obtained for rates up to 600 Lb P205/acre. Single applications of high rates of preplant fertilizer P, either broadcast and incorporated or deep knifed in a band, have resulted in higher yields and projected higher profits than annual conventional (100lb P205/acre) rates. Supplemental S has not shown a response and response to K2O has only occurred in the third year.

Aluminum Tolerance QTL in Diploid Alfalfa

M.K. Sledge, J.H. Bouton, J. Tamulonis, G. Kochert, and W.A. Parrot
Department of Crop and Soil Sciences , and Department of Botany,
University of Georgia,
Athens, Georgia, 30602 USA

Aluminum toxicity is the major growth-limiting factor for crop production on acid soils, which comprise up to 40% of the world's arable lands, including much of the United States. Surface liming, used to correct soil acidity, is expensive and does not neutralize harmful levels of subsurface aluminum. Alfalfa has been shown to be sensitive to aluminum toxicity. but conventional breeding methods have made little progress in introducing aluminum tolerance. We report the identification of quantitative trait loci (QTL) controlling aluminum tolerance in diploid alfalfa. Probes from our existing alfalfa RFLP map were used to identify and map the QTL. An in vitro callus growth bioassay was used to select aluminum tolerant and aluminum sensitive genotypes. The aluminum sensitive ssp. coerulea genotype (440501-2) was crossed with an aluminum tolerant genotype from a ssp. coerulea PI (464724) to produce and F2 population. The parents were screened with 146 cDNA probes, and 58 were mapped to nine linkage groups. The F2 genotypic classes for each of the 58 RFLP loci were contrasted with means from the callus growth bioassay, using ANOVA. We also used Mapmaker-QTL to identify markers associated with aluminum tolerance. Four markers, UGAc044. UCAc053, UGAc141, and UGAc782, were found to be associated with aluminum tolerance. UCAc044 had the greatest effect, accounting for 15% (LOD 2.3) of the variation in aluminum tolerance. These markers will aid in the future introgression of aluminum tolerance from tolerant germplasm into cultivated alfalfa. We are currently working to isolate allele specific markers near the UGAc044 locus, using bulked segregant analysis and AFLPs. Markers specific for the aluminum tolerant allele can be converted to sequenced tag sites and used to quickly screen large numbers of plants for the tolerant allele. We have isolated 5 bands specific for the tolerant allele, and 7 bands that are specific for the sensitive allele.

Brummer, E. C., J. H. Bouton, and G. Kochert. 1993. Development of an RFLP map in diploid alfalfa. Theor. Appl. Cenet. 86:329-332.
Michelmore, R. W., T. Paran, and R. V. Kesseli. 1991. Identification of markers linked to disease-resistance genes by bulked segregant analysis: A rapid method to detect markers in specific genomic regions by using segregating populations. Proc. Natl. Acad. Sci. USA 88:9828-9832.
Parrot, W. A., and J. H. Bouton. 1990. Aluminum tolerance in alfalfa as expressed in tissue culture. Crop Sci. 30:387-389.
Lincoln, S. E. M. J. Daly, and E. S. Lander. 1993. Mapping genes controlling quantitative traits using MAPMAKER/QTL version 1.1: A tutorial and reference manual. Whitehead Institute, Cambridge, Mass.
Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. van de Lee, M. Hornes, A. Frijters, J. Pot, J.Peleman, M. Kuiper, and M. Zabeau. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res. 23:4407-4414.

Predicting Acid/Aluminum Tolerance In Alfalfa Genotypes with Hematoxylin Staining.
Kelley, Rowena Y., Powell, D.; Yang, G.; Glass, M. North Carolina A&T State University, Greensboro , NC., 27411

Alfalfa (Medicago sativa L.), one of the most important forage legumes in the United Stales,
has been recognized as an aluminum sensitive species . Hematoxylin staining pattern is a technique that
has been used to evaluate differences in root growth and stain uptake between wheat cultivars sensitive
or resistant to high aluminum concentrations. Thc hematoxylin staining pattern was the focus of this
study because the procedure is simple, rapid and uses minimal space and labor. The objectives of this
study were: 1. To determine if hematoxylin staining can be used for evaluating acid/aluminum tolerance
in alfalfa 2. To determine the proportion of aluminum tolerant individuals for various alfalfa cultivars
with use of hematoxylin staining. Ten different alfalfa cultivars werc used: Apollo, ARC, Foundation
Vernal, Shenandoah, Spreador 2, WL 311, Saranac, Saranac AR, Cimarron, and Cimarron VR Forty
seeds from each cultivar were planted in Metro-mix 360 for use as controls. Forty seeds of each type
were also stained in a solution of hematoxylin for a period of two days. After staining, the seedlings
were transferred to Metro-Mix 360 for twenty-one days. After twenty-one days, the plantlets were
transferred to Porter's soil, an 80% aluminum saturated soil with a pH of 4.5, and grown in the
greenhouse for sixty days. After sixty days, fresh and dry root and shoot weights were determined and
performance, based on biomass, was compared to the tolerance level predicted by hematoxylin stain.
This study showed that some individuals in each cultivar were tolerant to the high concentrations of
aluminum (80% saturation). Individuals which scored >2 showed similar results with the most
significant difference being the numbers of individuals found at the various stain levels.

Cis-elements and Trans-acting Factors Required for
the Regulation of Alfalfa Isoflavone Reductase

B. Miao and N. Paiva
Plant Biology Division,
The Samuel Roberts Noble Foundation, Inc.
P. O. Box 2180, Ardmore, OK 73402

We are using both plant cell cultures and transgenic plants to study the regulation of isoflavonoid phytoalexin biosynthesis. In the isoflavonoid biosynthetic pathway, isoflavone reductase (IFR) is specific for the synthesis of medicarpin, the major antifungal isoflavonoid phytoalexin produced by alfalfa in response to fungal pathogen attack. Low levels of IFR mRNA and medicarpin conjugates accumulate constitutively in alfalfa cell cultures and root system. However, following fungal elicitor addition, high levels of medicarpin accumulate following rapid transcriptional activation of the IFR gene. Organ-specific expression may also become a strong local response when the plant is subjected to the infection. To understand how the IFR gene is regulated, the IFR promoter has been analyzed both in vitro and in vivo .
By using promoter fragments in gel retardation assays (GRAs), a specific DNA-binding complex was observed by using nuclear extracts from both elicited and unelicited cell cultures. The level of binding activity increases about 5-fold within 90 min of elicitation. At later time points, the signal decreased rapidly, indicating the response to elicitation is rapid but transient. Further detailed GRA analysis both indicated that an AT-rich region from -270 to -130 bp upstream of the start of transcription is essential for transcription factor binding. In this same AT-rich region, DNase I footprinting revealed two elicitor induced regions, Box I and Box II. Both sites are about 20 bp long and AT-rich with very similar DNA sequences. The functions of these elements were tested in transgenic alfalfa plants using a series of IFR promoter deletion-GUS fusion constructs. Deletion of Box I and Box II appears to greatly reduce the expression of the GUS reporter gene in vivo. By using nuclear extracts, the protein blots were probed with either IFR fragment B or the synthetic multimer including consensus sequence of Box I and Box II. The Southwestern blot analysis revealed a 50-60 kD protein band. Phosphatase treatment eliminates the DNA binding activity of nuclear extracts, suggesting that the factor may be present at all times, then activated via phosphorylation by a protein kinase element of a signal transduction pathway.
References 1. Oommen, A., Dixon, R. A., Paiva, N. L., 1994. The elicitor-inducible alfalfa isoflavone reductase promoter confers different patterns of developmental expression in homologous and heterologous transgenic plants. Plant Cell 6: 1789-1803.

Paiva, N. L., Edwards, Y. Sun, Y., Hrazdina, G., Dixon, R. A., 1991. Stress responses in alfalfa (Medicago sativa L.). 11. Molecular cloning and expression of alfalfa isoflavone reductase, a key enzyme of isoflavonoid phytoalexin biosynthesis. Plant Mol. Biol. 17: 653 -667.

Toward Metabolic Engineering of Alfalfa for Production of Prenylated
Isoflavonoid/Pterocarpan Phytoalexins

Z. Guo and N. L. Paiva
Plant Biology Division, The Samuel Roberts Noble Foundation,
2510 Sam Noble Parkway, Ardmore, Oklahoma 73402, U. S. A.

Upon pathogen infection, alfalfa (Medicago sativa L. ) accumulates the simple pterocarpan phytoalexin medicarpin as a defense response. Some legumes, such as beans (Phaseolus vulgaris), soybean ( Glycine max), and white lupin (Lupinus albus), produce prenylated isoflavonoids and/or pterocarpans either upon elicitation or in a constitutive manner (1). In contrast, no prenylated isoflavonoids have been reported to accumulate in alfalfa. It has been shown that prenylated isoflavonoids and pterocarpans are more fungitoxic than their non-prenylated counterparts. Also, prenylated isoflavonoids were shown to inhibit feeding of insect herbivores. Aiming at improving fungal and pest resistance, we are exploring the feasibility of altering the phytoalexin profile in alfalfa by introducing a suitable exotic prenyltransferase. We focused on the bean system as a source because of the high structural similarity between its prenylation substrate and medicarpin and because of a previous report that bean prenyltransferase is active on medicarpin (2).
Using a microsomal preparation from CuCl2-treated bean as catalyst, we demonstrated stereospecific biosynthesis of 10-dimethylanylmedicarpin from dimethylanyl diphosphate and (-)-medicarpin. Structural verification of the reaction product was achieved using chromatographic and W, MS and lH-NMR spectral analyses. We are currently performing bioassays to determine the fungitoxicity of this product relative to medicarpin. Medicarpin prenylation activity appeared 4-8 hours after CuC12 induction; it increased and reached the maximum level in about 32 hours. The accumulation of the phytoalexins phaseollin and kievitone followed a similar pattern. Earlier reports indicated that the pterocarpan prenyltransferase is associated with chloroplast membranes, and so we developed a protocol for the large scale preparation of chloroplast envelope membranes (CEM) from bean seedlings that are treated with CuCl2 to induce high levels of prenyltransferase activity. We were able to show that this preparation had specific prenylation activity 150 fold higher in comparison with microsomal fractions. Further purification (double the specific activity) was achieved by differential solubilization. Rabbit antisera against partially purified CEM proteins, both native and denatured were generated and used to screen a bean cDNA expression library. A total of 144 positive clones have been isolated. Work is underway to characterize these clones according to CuCl2-inducibility and sequence similarity to known prenyltransferases.

References
1. Biggs, D. R., Welle, R., and Grisebach, H. 1990. Intracellular localization of prenyltransferases of isoflavonoid phytoalexin biosynthesis in bean and soybean. Planta 188:244-248.

2. Biggs, D. R., Welle, R., Visser, R. F., and Grisebach, H. 1987.
Dimethylallylpyrophosphate:3,9-dihydroxypterocarpan-10-dimethylallyl
transferase from Phaseolus vulgaris. FEBS Letters 220:223-226.

Alfagene.s - A Comprehensive Internet-accessible ACE Database for Alfalfa
http://probe.nalusda.gov

D. Z. Skinnerl,2, P. C. St. Amand1,2, and S. M. Ramaiah2
USDA-ARS and Agronomy Department, Kansas State University
Manhattan, Kansas, 66506-5501, U.S.A.

A comprehensive computer database of the alfalfa genome has been created using the powerful "ACE" (1) software. The goal of this database is to make all known information relevant to the alfalfa genome accessible from a single location. Currently, the database contains information in 17 classes (Table 1). Data have been compiled from numerous sources, including the USDA Germplasm Resources Information Network (GRIN), The Certified Alfalfa Seed Council, and hundreds of published works. The data are inter-related such that information in one data class can lead immediately to information in another class simply by following a hypertext link. For example, a DNA sequence that is cited in a literature citation can be displayed immediately simply by selecting the name of the sequence from within the literature citation. Numerous powerful search and analysis tools are available within the database. For example, DNA sequences may be fully analyzed with resident tools that can generate restriction endonuclease maps, or search for open reading frames. Graphic images of linkage groups can be displayed, and recombination frequencies are available. Graphic images of metabolic pathways are available, selecting any part of the pathway displays complete information on that part. Search tools can be applied to any part of the database; for example, literature citations may be searched for keywords or specific authors. The database is accessible on the world-wide-web at: http://probe.nalusda.gov .

Reference
1. Richard Durbin and Jean Thierry Mieg (1991-). A C. elegans Database. Documentation, code and
| data available from anonymous FTP servers at lirmm.lirmm.fr, cele.mrc-lmb.cam.ac.uk and
ncbi.nlm.nih.gov.

Risk of Alfalfa Transgene Dissemination and Scale Dependent Effects

P. C. St. Amandl, D. Z. Skinner1, and R. N. Peaden2
lUSDA/ARS and Agronomy Department, Kansas State University, Manhattan KS
66506 and USDA/ARS, Irrigated Agricultural Research Center, Prosser WA 99350

Concerns have been raised over the release of genetically engineered plants into the environment. Because pollen can readily disseminate engineered genes throughout a plant population or into a related species, the risk of inadvertent dispersal of engineered genes must be known. Measures of that risk include the rate of pollen-mediated gene spread, the maximal dispersion distance of pollen, and the spatial dynamics of pollen movement within seed production fields; none of which are known for alfalfa (Medicago sativa L.), ïan insect pollinated species. Lacking from most risk assessment studies is the investigation of scale dependencies of the data. Using a rare, naturally occurring molecular marker, pollen dispersal can be tracked without introducing engineered genes into the environment. Using PCR-based methods, a suitable marker was found in an intron of the alfalfa glutamine synthetase gene. This marker was not present in a broad range of alfalfa genotypes, and was used to measure gene dispersal within seed production fields. Preliminary results indicate that bees (Megachile spp.) used in commercial seed production show a bi-directional bias when pollinating, resulting in the movement of pollen from marker plants directly toward and away from the colony hive. Gene movement within fields from individual plants occurs only over very short distances, up to 4 m in this study. By examining widely-dispersed, individual escaped alfalfa plants and their progeny using RAPD markers, gene movement among escaped alfalfa plants has been confirmed for distances up to 230 m. Progeny from escaped plants at greater distances have been collected, but have not yet been evaluated. Long-range dispersal of genes from alfalfa seed and hay production fields has been confirmed for distances up to 1000 m using small trap plantings of alfalfa (Fig. 1). A novel field design allowed an accurate investigation of scale dependencies on gene movement by comparing dispersal from research scale plots with that of commercial scale fields. Scale effects were large. Gene markers moved from research scale plots a maximum of 300 m, while markers from near commercial sized fields moved up to 1000 m and were present in 22.2% of the progeny at that distance (Fig. 1). This work shows that genes can be dispersed great distances from alfalfa fields and is not related to the short range dispersal that occurs within a field. The probability of long-range transgene dissemination from commercial fields is high, as is the probability for dissemination among individual escaped alfalfa plants.

Mapping of Simple Sequence Repeats (SSR) DNA Markers in Diploid
and Tetraploid Alfalfa

Noa Diwan1,2, Joseph H. Bouton3, Gary Kochert4, Arvind A.
Bhagwat1,2, and Perry B. Cregan1.

lSoybean and Alfalfa Research Lab., USDA-ARS, Beltsville, MD.
20705. Department of Agronomy, University of Maryland, College
Park, MD 20742. 3 Department of Agronomy, University of Georgia,
Athens, GA 30602. Department of Botany University of Georgia,
Athens, GA 30602.

Cultivated alfalfa (Medicago sativa) is an autotetraploid. However, all three existing alfalfa genetic maps resulted from crossing of diploid alfalfa. A study was undertaken to evaluate the use of Simple Sequence Repeat (SSR) for mapping diploid and tetraploid alfalfa. Ten SSR loci were incorporated into an existing F2 diploid alfalfa RFLP map, and were also mapped in an F2 tetraploid population. The tetraploid population had two to four alleles in each of the loci examined. Following the segregation of these alleles in the tetraploid mapping population generally was clear and easy. Linkage relationships at the tetraploid level were determined by using a single dose allele
(SDA) analysis, where each allele is scored as a dominant marker independently of the other alleles at the same locus. The SDA diploid map was also constructed to compare mapping using SDA to the standard co-dominant method. Linkage groups were generally conserved among the tetraploid and the two diploid linkage maps, except for segments where severe segregation distortion was present. Segregation distortion which was present in both tetraploid and diploid populations probably resulted from
inbreeding depression. Therefore, a better approach to map
alfalfa should be in a non-inbred mapping population such as F1.
The ease of analysis together with the abundance of SSR loci in
the alfalfa genome indicated that SSR markers should be a useful
tool for mapping in tetraploid alfalfa.

References
Brummer, E.C., J.H. Bouton, and G. Korchet. 1993. Development of
an RFLP map in diploid alfalfa. Theor Appl. Genet. 86:329
332.
Cregan, P.B., A.A. Bhagwat, M.S. Akkaya, and Jiang Rongwen.
1994. Microsatellite fingerprinting and mapping of soybean.
Methods. Mol. Cell. Biol. 5:49-61.
D. A. Silva, J. A. G., M. E. Sorrells, W. L. Burnquist, and S. D.
Tanksley. 1993. RFLP linkage map and genome analysis of
Saccharum spontaneum. Genome, 36:782-791
Yu, K. F., and K. P. Pauls. 1993. Segregation of random amplified
polymorphic DNA markers and strategies for molecular mapping
in tetraploid alfalfa. Genome, 36:844-851.

Divergent Selection for Self-fertility in Alfalfa and Correlated Responses for Forage Yield

D. Rosellini, F. Veronesi and G. Barcaccia
Istituto di Miglioramento Genetico Vegetale, University of Perugia
Borgo XX Giugno 74, 06100 - Perugia, Italy

Different breeding strategies are being followed in advanced alfalfa breeding. The role of recurrent selection in population improvement has been recently reexamined in the light of the greater importance attributed to 'complementary gene interactions' with respect to allelic interaction in determining vigor. According to this interpretation, inbreeding in the selection process could help reduce the genetic load and fix favorable linkage groups. A different breeding strategy aims at developing double cross cultivars based on well combining, self-sterile clones through the somatic seed technology. Self-fertility in alfalfa has been widely studied: it shows continuous variation in the populations, is associated to cross-fertility and is drastically reduced by inbreeding, usually more than forage yield. Even if a gametophytic self-incompatibility system has been suggested, the cause of self sterility seems to be a 'relational incompatibility' that can be related to genetic load. If the genetic loads for fertility and for forage yield are at least partly coincident, selection for self-fertility may help reduce the genetic load and bring about positive correlated responses for forage yield. In this perspective, selecting SS genotypes for developing hybrids would not be a valuable strategy. Divergent selection for self-fertility was applied in a central Italian alfalfa landrace with the purpose of developing materials to test the previous hypotheses. Four hundred and fifty three plants were grown in pots under bee proof cages and evaluated for self-fertility (seeds produced per floret tripped, 30 to 40 florets per plant); 10 self-fertile (SF) and 10 self-sterile (SS) plants were selected on the basis of 3 evaluations of self-fertility made in different environmental conditions. The 10 clones of each group were hand crossed without emasculation according to a n(n-l) diallelic design; equal amounts of seeds from reciprocal crosses were pooled, thus obtaining 45 full-sib families per group. Ten plants per family were grown in pots randomized within bee-proof cages and evaluated for self-fertility. The following spring, 5 random plants per family were evaluated for green matter yield. Selection was effective for self-fertility (h2R=0.52), but not for self-sterility. Such results are expected if self-sterility is due to genetic load (figure 1). The SF plants, on the contrary, would have a low genetic load, i.e., a high frequency of dominant genes for fertility, and their progenies would show improved self fertility. This hypothesis is consistent with the low cross-fertility of the SS plants observed in the diallel cross, when compared to that of the SF clones. The forage yield of SS progenies (21.7 g per plant) was significantly higher than that of SF progenies (13.6), which did not differ from the control (13.22). This could be due to a lower than normal percentage of Sl plants in the SS progeny. A dense stand trial is underway to verify whether the forage yield superiority of SS progeny is mantained in competitive conditions, in which Sl plants are selected against. Inbred SF materials are also being developed to test the effect of deliberate selection for self-fertility during inbreeding on vigor.

Ploidy Determination of Medicago sativa ssp. falcata germplasm:
Step 1 in a falcata germplasm enhancement program

E. C. Brummer and S. Sin
Agronomy Department, Iowa State University
Ames, IA 50011 USA

Alfalfa, Medicago sativa, breeding programs in most of the United States have traditionally concentrated on germplasm containing predominantly ssp. sativa. M. sativa ssp. falcata has been used only to a limited extent, mainly for northern and rangeland areas due to its generally superior winter- and drought tolerance. For other traits such as forage yield and disease resistance, falcata material is inferior to .sativa. Molecular studies have shown that falcata germplasm is genetically distinct from sativa (Kidwell et al. 1994, Brummer et al. 1991) indicating that it may include desirable alleles for various agronomic traits. Bingham et al. (1995) contended that alfalfa heterosis is due to the presence of complementary linkage blocks. However, Barnes et al. (1977) pointed out that the current methods used to breed alfalfa may be breaking apart desirable blocks, bringing genes and alleles controlling important traits into equilibrium. Thus, the addition of new linkage blocks may be necessary to allow continued improvement in alfalfa cultivars and enhanced falcata germplasm is one possible source of these blocks.
The long term objective of this project is to develop falcata germplasm that may be useful for alfalfa breeding programs. The current germplasm collection consists of mostly unimproved materials, which have not seen several decades of concerted breeding efforts as has sativa. The specific objective of this research is to determine the ploidy level of all available falcata acccssions. We plan to use the tetraploid accessions in further breeding work.
Ploidy has been determined on 159 falcata accessions to date. All materials were obtained from the Western Regional Plant Introduction Station at Pullman, WA. Chromosome counts were made on root tip squashes from young seedlings grown in petri plates. Five plants per accession were counted; a minimum of 10 cells per plant were observed. Approximately 63% of the accessions counted were tetraploid, 33% diploid, and 6% mixtures of diploids and tetraploids.
Enhancing the falcata germplasm pool should be conducted at the tetraploid level ir the material is intended to be introgressed into current alfalfa breeding material. Thus, determination of the ploidy of each accession is useful prior to planting selection nurseries. Actual chromosome counts are the best way to assess ploidy because morphological differences are not always reliable indicators. The fact that almost two-thirds of the falcata accessions for which seed is available are tetraploids will allow us a large diversity of material from which to begin selection.
Although the enhancement of this germplasm will not be easy or its results quickly applied to current cultivar breeding programs, without this breeding work little use will ever be made of the falcata collection. This falcata enhancement project will use the identified tetraploid accessions as the basis from which to select for various agronomic traits, including winterhardiness, upright growth, leafiness, and later fall dormancy. Even though falcata germplasm introgression into traditional hay cultivars may be limited, this material may find a larger niche in cultivars developed specifically for long-term permanent pastures, where intensive cutting and management systems are of less importance. A shift in the paradigm of alfalfa breeding may result in further use of enhanced falcata germplasm.
References:
Barnes, D.K., E.T. Bingham, R.P. Murphy, O.J. Hunt, D.F. Beard, W.H. Skrdla, and L.R. Teuber. 1977.
Alfalfa germplasm in the United States: Genetic vulnerability, use, improvement, and maintenance.
USDA Tech. Bull. No. 1571. USDA-ARS, Washington, D.C. 21 p. Bingham, E. T., R. W. Groose, D. R. Woodfield, and K. K. Kidwell. 1994. Complementary gene interactions in
alfalfa are greater in autotetraploids and diploids. Crop Sci. 34:823-829. Brummer, E. C., G. Kochert, and J. H. Bouton. 1991. RFLP variation in diploid and tetraploid alfalfa Theor.
Appl. Gen. 83:89-96. Kidwell, K. K., D. F. Austin, and T. C. Osborn. 1994. RFLP evaluation of nine Medicago accessions
representing the original germplasm sources for North American alfalfa cultivars. Crop Sci. 34:230-236.

Progeny Test Based on Morphological and Molecular Markers to Verify the Occurrence of
Parthenogenesis in an Apomeiotic Mutant of Diploid Alfalfa

G. Barcaccia1, S. Tavoletti2, M. Falcinellil and F. Veronesi2

(1) Istituto di Miglioramento Genetico Vegetale, University of Perugia (Italy);
(2) Dipartimento di Biotechnologie Agrarie ed Ambientali, University of Ancona (Italy).

In the polysomic polyploids, such as cultivated alfalfa, maximum heterosis may be expressed by a few elite individuals of the population but not by the entire population. An approach to perpetuate the elite individuals and preserve heterosis over generations could be the use of apomixis. The apomictic reproduction consists of two essential processes: apomeiosis (formation of eggs avoiding meiotic reduction) and parthenogenesis (development of embryos without fertilization). Thus, apomixis has the potential of cloning plants through the seed and it furnishes a unique opportunity for developing superior tetraploid cultivars with permanently fixed heterosis. Apomixis as a whole has not been detected in the genus Medicago, but components of apomixis are present. The formation of unreduced eggs through apomeiosis in a diploid plant of M. sativa subsp. falcata (L.) Arcang., named PG-F9 is an extremely interesting feature of apomixis, as is the induction of haploid parthenogenesis in tetraploid lines of alfalfa, which has been widely exploited to obtain CADL populations. Our goal is to eventually combine all the components of functional apomixis in alfalfa. Studies conducted in diplosporous species report that apomeiosis and parthenogenesis could be processes tightly associated. Information on the genetic composition of progenies from the mutant PG-F9 cannot be obtained on the basis of cytoembryological analyses. Therefore, a progeny test based on morphological and molecular markers was carried out to verify the parthenogenetic capability of apomeiotic eggs. Four morphological traits such as leaf shape, stipule form, stem pigmentation and flower color showed to be effective in the preliminary screening of progenies and most of the plants were classified as aberrants (from sexual reproduction). Molecular investigations by means of RAPD fingerprinting and RFLP marker detection conducted in the progenies morphologically classified as maternal allowed two plants genetically similar but no identical to PG-F9 to be discovered. Owing to the high number of molecular markers conserved as in PG-F9 and because the great discriminating efficiency of primers and probes used, these plants could likely be generated through parthenogenesis of apomeiotic eggs. According to the degree of similarity distribution in the self progeny, the differences found at few genomic loci could be explained by an origin from selfing only assuming a non-ordinary conservation of the maternal genotype. In fact, considering the PG-F9 is not inbred, there was a remote possibility to find a plant from selfing identical to PG-F9 by using the 36 selected ten-mer primers (that generated in PG-F9 a total of 269 RAPD bands) and by investigating the 12 heterozygous RFLP loci. The presence of DNA rearrangements could be explained by the occurrence of FDR mechanisms, which were not observed at the cytological level, whereas an autosegregation process should not be involved since the metaphase analysis revealed the normal somatic chromosome number. The overall judgment about the sexual or apomictic origin of the plants morphologically e molecularly similar to the apomeiotic mutant PG-F9 will require further investigations at the nucleotidic level.

Assessment of genetic variability for cryoprotective oligosaccharide accumulation in cold acclimated leaves of alfalfa.
Y. Castonguay, P. Nadeau & R. Michaud. Agriculture and Agri-Food Canada,
Soils and Crops Research and Development Centre, Sainte-Foy, Quebec, Canada, GlV 2J3.

Cold acclimation of alfalfa is associated with marked changes in the carbohydrate composition of overwintering crowns and roots (McKenzie et al., 1988). We have recently shown that differences in freezing tolerance between winterhardy and nonhardy cultivars (cvs.) are more closely related to crown levels of the oligosaccharides of the raffinose family (RFO; stachyose and raffinose) than to the levels of sucrose (Castonguay et al., 1995). This observation suggests that cryoprotective RFO could constitute a potential trait to screen for cold tolerance in alfalfa. However, the extent of genotypic variation for RFO accumulation within populations of alfalfa and its potential for the genetic improvement of cold tolerance of alfalfa needs to be further investigated. The objectives of this study were to assess whether RFO accumulate in leaves of alfalfa and to determine the extent of the variability of leaf RFO within an alfalfa cv. This information is important in order to develop a non-destructive method to screen alfalfa genotypes for their potential to accumulate cryoprotective oligosaccharides.

We found that raffinose accumulates in cold-acclimated leaves whereas stachyose remained at very low levels. Cold-induced accumulation of raffinose gradually increases from non-hardy to very hardy cvs. Superior levels of leaf raffinose in plants acclimated at a nonlethal subfreezing temperature relative to plants acclimated at a low non-freezing temperature further indicate that leaf raffinose is closely linked to freezing tolerance in alfalfa. Analysis of leaf raffinose in 497 genotypes of the cv. DK-125 showed a wide distribution varying from nearly 0 up to 20 mg g-1 dry weight raffinose with a mean skewed toward lower values (Fig. 1). We observed a contrasting accumulation of leaf raffinose between cold acclimated cuttings from genotypes selected either for their high or low level of leaf raffinose. The extent of the variability within each genotype indicates that the variation in cold-induced accumulation of leaf raffinose within a population is genetically determined. Our results suggest that cold-induced accumulation of leaf raffinose could potentially be used as a non-destructive screening method to improve cold tolerance in alfalfa.
References

Castonguay, Y., Nadeau, P., Lechasseur, P. and L. Chouinard. 1995. Differential accumulation of carbohydrates
in alfalfa cultivars of contrasting winterhardiness. Crop Science 35: 509-516.

McKenzie, J. S., R Paquin, and S. H. Duke. 1988. Cold and heat tolerance. P. 259-302. In A. A. Hanson et al.
(ed.) Alfalfa and alfalfa improvement. Agron. Managr. 29. ASA, CSSA, and SSSA, Madison, WI.

Analysis of Seed Storage Proteins in Regenerated Plants of Alfalfa
(Medicago varia Mart.)

O. Tzeveleva., S. Peltek, E. Deineko
Institute of Cytology and Genetics, Novosibirsk, Russia

Different biotechnological approaches can be utilized for inducing more variable cross pollinated species (for instance, tetraploid alfalfa). One of them is based upon the phenomenon of somaclonal variation. Alfalfa somaclones <Ire widely used as a source of variation for basic research and breeding. It is often impossible to confirm the genetic nature of this phenomenon. To understand the origin of somaclonal variation, we found it most appropriate to use seed storage proteins, as a part of them are largely under monogenic control, they are easy to identify, and allow a large number of phenotypes to be analyzed at a time. The experimental material on which we studying changes in the pattern of seed storage proteins in alfalfa regenerants (Medicago varia Mart.) was obtained as indicated in Fig.1. The patterns were analyzed on the self-pollinated progeny of all the three donor regenerant pairs after SDS- PAAG. About 100 seeds of each of the plants were analyzed individually. In total, 38 individual components have been revealed. We counted the number at which each of the 38 components occurred in the patterns. If each component were under monogenic control, the expected frequencies would be as follows: AAAA-1; AAAa - 0,96 ;AAaa -0,92 ;Aaaa - 0,76 and aaaa - 0. The individual components of the seed storage protein patterns of the regenerant (R) and the donor (D) were divided into two groups: one with those occurring as expected or so, and one with those occurring significantly otherwise. A variety of genetic factors (incompatibility systems, selection among gametes etc.) might account for the observed difference in frequency. Further analysis was performed on the former group (Table 1.) and the difference between the respective frequencies of the individual components within each pair of plants was proved significant. We did not observed new (mutant) variants of a component, which allows us to conclude that tissue culture conditions must be influencing only the expression of genes controlling the synthesis of these components.

Callus Initiation and Development of Medicago sativa L. Cultured in vivo

Guochen Yang and Marihelen Kamp-Glass
Department of Natural Resources and Environmental Design
North Carolina A&T State University
Greensboro , NC 2741 1

Plant transformation techiques have been successfully used to incorporate particular genetic constructs to plant genome to improve disease and/or stress resistance/tolerance for many plant species, such as alfalfa (Kuchuk et al., 1990). In-vitro shoot regeneration protocol must be first established and optimized in order to establish an efficient and reliable transformation system. As a member of legume family, alfalfa is very difficult for tissue culture, although success has been achieved on embryogenesis (Meijer & Brown, 1987). However, limited research was done on plant growth regulators' (PGR) effects on alfalfa callus initiation and development in order to enhance embryogenesis and/or organogenesis (Saunders & Bingham, 1972 and 1975). Therefore, this research was initiated with following objectives: (a) to evaluate how different PGR influence alfalfa callus initiation and development; (b) thus to establish a foundation for further development of shoot organogenesis protocol to assist transformation.
Seeds of cultivar WL 31 1 were used as explants. Seeds were sterilized by soaking in 15% (0.78% NaOCl) bleach solution for 15 minutes, rinsed 3 times with sterile distilled deionized water. M.S. basal medium supplemented with 3% sucrose and 0.7% agar, pH 5.8, plus BA at 1, 5 or 10 mg/l; zeatin at 0.1, 1, 2, or 5 mg/l; 2,4,5-T at 0. 1, 1, 5, 10 mg/l as treatments. Four (4) sterilized seeds were transferred to each petri dish containing approximately 25 ml of medium of particular treatments. Petri dishes were placed in the culture shelves with 16 hours of light provided by cool whte fluorescent tubes at 23 + 3 °C.
BA at I or 5 mgll enhanced callus initiation and development toward shoot organogenesis. The calli regenerated from these two treatments are compact, solid and dark green. SNOW white calli were also noticed. BA at 5 mg/l produced more sucll calli than at I mg/l. However, BA at 10 mg/l inhibited callus development, only small amount of watery calli was produced.
Zeatin at 5 mg/l was the best among the concentrations tested. Large amount of dark green, compact and solid calli was regenerated while other treatments only produced watery and soft calli. At 5 mg/l, semi-compact, green, bright purple, and snow white calli were also developed.
Concentration of 2, 4,5-T at 1 mgll promoted callus production, while higher concentrations inhibited callus initiation. However, only soft calli with sometimes snow white spots were developed from 2,4,5-T treatments. A cytokinin is apparently needed to stimulate shoot organogenesis from such calli.
Calli regenerated can be further stimulated to induce shoot initiation, thus to establish a regeneration protocol to assist genetic transformation.

MOLECULAR GENETICS OF A MODEL-PLANT: MEDICAGO TRUNCATULA

T.Huguet (l), M.Gherardi (1), L.Tirichine (1),
I. Bonnin (2), G. Genier (2), J. Ronfort (2), J.M.Prosperi (2)
(1) LBMRPM, CNRS-INRA, BP27, 31326 Castanet-Tolosan Cedex, France.
(2) SGAP, INRA, Domaine de Melgueil, 34130 Mauguio, France.

Alfalfa (Medicago sativa) is a forage crop of considerable worldwide agronomic importance having a high nitrogen fixation and protein productivity potential. Unfortunately, both fundamental and applied research on alfalfa is limited by its complex genetic structure (autotetraploid, allogamous and with inbreeding depression). For these reasons, we use the mediterranean forage legume Medicago truncatula as a model-plant to circumvent the limitations of alfalfa and to combine genetic and molecular biological studies (Barker et al. 1990). M.truncatula is an annual forage, diploid, autogamous plant of low DNA content and with which transgenic plants can be routinely obtained. M.truncatula is nodulated by well-known Rhizobium meliloti strains and is of good agronomical value for mediterranean regions where they are utilized to improve the soil structure, increase soil nitrogen and as a source of winter forage. Selection of medics for introduction in mediterranean marginal lands started in 1985 at INRA France (Prosperi 1993). A high level of polymorphism can be observed both between and within natural populations for as well neutral PCR markers (RAPD) as morphological traits Bonnin et al 1996 a,b). This variability is used in breeding programs to improve characters such as tolerance to cold, to diseases or to grazing.
In order to identify and isolate plant genes related to Rhizobium-legume symbiosis and/or of agronomical interest, we are engaged in the construction of a genetic map of M.truncatula. This map is based on crosses between 4 polymorphic genotypes originating from geographically different areas: 1 natural population from France, 2 from Algeria and l Australian cultivar (Jemalong). Using three of these crosses, recombinant lines have been obtained by single seed descent. From a single cross between Jemalong and a natural population, we have already been able to map 50 RAPD markers, 2 isoenzymes (PGD, PGM) and 6 known genes (rDNA, ENOD12, MTGSa, MTGSb, MTLbl, MTLec2). This map defines 9 linkage groups (2n=16) spanning over 571 cM. Haploid DNA content was measured by flow cytometry to be 0.58pg/lC, ensuring approximately lMB/cM. We are presently mapping a set of symbiotic mutations (Nod++, Nod- and Fix-) obtained by exposure to gamma rays (Sagan et al 1995). In addition, M. truncatula is known for a long time to show a strong cultivar x Rhizobium strain specificity. The above natural populations are now screened with a set of wild-type Rhizobium meliloti bacterial strains.
This genetic map will be a powerful tool for the positional cloning of symbiotic genes and for the breeding of mediterranean forage legumes as well as for studying evolutionary processes within Medicago genus.
References

Barker, D. G., Bianchi, S., Blondon, F., Dattee, Y., Duc, G., Flament, P.,Gallusci, P., Genier, P., Guy, P., Muel, X., Tourneur, J., Denarie, J., and Huguet, T. 1990. Medicago truncatula, a model plant for studying the molecular genetics of the Rhizobium-legume symbiosis. Plant Mol. Biol. Rep. 8, 40-49.
Prosperi, J. M., 1993. Selection of annual medics for French Mediterranean regions. In Workshop on introducing the ley farming system to the Mediterranean basin. 173-191. (1989) Perugia (Italy). Eds.S. Christiansen, L. Materon, M. Falcinelli et P. Cocks.
Sagan, M., Morandi, D., Tarenghi, E., Duc, G. 1995. Selection of nodulation and mycorrhizal mutants in the model plant Medicago truncatula Gaerth after gamma rays mutagenesis. Plant Science,111,63-71.
Bonnin, I., Huguet, T., Gherardi, M., Prosperi, J. M., and Olivieri, I. 1996 a. High level of polymorphism and spatial structure in a selfing plant species, Medicago truncatula (Leguminosae), using RAPD markers. American Journal of Botany. 83(6) In press.
Bonnin, I., Prosperi, J. M., and Olivieri, I., 1996 b. Genetic markers and quantitative genetic variation in Medicago truncatula (Leguminosae): a comparative analysis of population structure. Genetics. In press.

Evaluation of cold tolerance in annual medics
with Potential for use in rotation with wheat on the U.S. High Plains

J. M. KRALL, R. H. DELANEY, D. A. CLAYPOOL and R. W. GROOSE

Department of Plant, Soil and Insect Sciences, University of Wyoming,
PO Box 3354, Laramie WY 82071 USA

On the U.S. Western High Plains winter wheat is most often grown in a 2 yr rotation that involves a 14 mo fallow period. Inclusion of a self-regenerating winter annual legume as "green fallow" could contribute to a more economically and ecologically sustainable agroecosystem. Based on geographical origin in cold climates, we identified 66 potentially winter-hardy experimental lines representing 11 annual Medicago spp. These lines were tested along with 7 commercial Australian annual medic cultivars, 2 winter-hardy alfalfa cultivars ('Ladak 65' and 'Anik'), and 1 black medic cultivar ('George') in trials seeded in Aug 1995 at three locations in Wyoming. M rigidula and rigiduloides proved to be the most winter-hardy annual medic species with 74% and 56% of lines, respectively, surviving at one or more locations in June 1996 (Table 1). Limited survival of M. orbicularis (20% of lines) was observed. No experimental lines or cultivars of M. aculeata, arabica, disciformis, littoralis, polymorpha, rugosa, scutellata, tornata and truncatula survived at any location. At Laramie, all surviving lines were flowering on 10 June except M. orbicularis and M. sativa lines. Very low survival (Ladak 65 only) in sandy soil at Torrington was due to desiccation in late winter when adjacent winter wheat plants were also winter killed.

Breeding Medicago lupulina for Pasture and Green Manure

G. V. Stepanova, E. N. Atlakova, G. P. Zatchina
William Fodder Research Institute, Lugovaja, 141740, Russia

Black medic (Medicago lupulina) - the most widespread poli morphic species of genus Medicago, is growing in different climatic conditions. There are spring- and winter, annual- and biannual forms. It is self-pollinating. In Russia in the 19th century black medic was considered to be one of the best pasturable plants. Then it have been forgotten for many years, and only since 1989 selection work with this valuable fodder crop began in William Fodder Research Institute. The main selected signs are: yield of the dry matter, seeds, plant regrowth, winterhardiness, the high symbiotic nitrogen fixation ability, resistance to the main diseases. About 100 wild growing specimens have been used as an initial material. The methods of breeding are: polyploidy, mutagenesis, directional selection. Black medic samples of pasture type have been tested in grass mixture with early spring sowing. While medic samples has been used as green manure it was inoculated with appropriate Rhizobium strain, sowed in August and ploughed in the following year.

The dry matter harvest of pasture mixture with different breeding medic samples varied from 13,13 to 15,26 t/ha, with medic content from 7,83 to 13,58 t/ha for two years using. Winter hardiness of the best two samples was 80 and 87%, others samples-4 25%. Black medic can grow in the pasture for many years owing to self-sowing. When black medic was used as green manure, it was cut at the end of June, then it regrown and in September or in October it was ploughed. The dry matter harvest of testing samples in first cutting was 3,0-3,5 t/ha. Content of nitrogen in the roots and regrowth was 137-201 kg/ha. Breeding to increase nitrogen fixation ability and resistance to main diseases was held in order to receive large harvest of dry matter and seeds. Six specifical for black medic Rhizobium meliloti strains have been created. The inoculation increased dry matter harvest by 53,7-72,2%, seeds - 112,2-149,6%. The main disease, striking black medic is root rot, caused by fungus of genus Fusarium. Species F.culmorum, F. oxysporum, F. moniliformi, F.solani, F. sambicium are the most aggressive for M.lupulina. The best breeding samples have been struck by 17,7-20,0%, the others 26,4-45,1%.

Assignment of Medicago rigidula accessions in the NPGS
into the two species M. rigidula and M. rigiduloides

D.C. HEFT and R.W. GROOSE

Department of Plant, Soil and Insect Sciences, University of Wyoming,
PO Box 3354, Laramie WY 82071 USA

Recently the cold-tolerant annual Medicago species M. rigidula was differentiated into the two species M. rigidula (L.) All. and a new species, M. rigiduloides E. Small, based on number of coils per pod and pod spine curvature where European types (M. rigidula) have, on average, fewer coils per pod and more hooked pod spines than Asian types (M. rigiduloides; Small et al. 1990, Can. J. Bot. 68:2607; Small 1990 ibid.:2614). Small et al. (1990) indicate that these species are even more distinctly differentiated by pollen morphology where M. rigidula has cylindrical pollen in contrast to the pyramidal-triangular pollen of M. rigiduloides. Recent research at the South Australian Research and Development Institute and the University of Wyoming has demonstrated differences between these species in Rhizobium specificity and winter-hardiness (Groose et al. 1996; Krall et al. 1996; see abstracts in this volume). Groose et al. (1996) have also demonstrated a partial barrier to interspecific fertility between M. rigidula and M. rigiduloides. The goal of this study was to differentiate all available accessions of "M. rigidula" in the U.S. National Plant Germplasm System into M. rigidula and M. rigiduloides. Three plants of each line were grown in the greenhouse. Pollen was observed at flowering and pods traits were evaluated at maturity according to the key of Small et al. (1990). Lines were classified as M. rigidula and M. rigiduloides on the basis of cylindrical versus pyramidal-triangular pollen, respectively. For lines that produced unusual pollen phenotypes or appeared to be mixed lots, pollen of 10-20 additional plants were observed. In total, 307 ofthe 332 accessions of"M. rigidula" in the NPGS were examined. As summarized in Table 1, the vast majority of accessions produced either predominately cylindrical pollen (M. rigidula, 103 lines total) or predominately pyramidal-triangular pollen ( M. rigiduloides, 192 lines total). Seven accessions produced pollen typical of both species, as well as some pollen grains of an intergrade phenotype. These accessions were classified as 'intermediate.' Finally, five accessions proved to be mixed lots. Four Asian accessions included plants of both M. rigidula and M. rigiduloides and one European accession included plants of M. rigidula and M. polynmorpha. The intercontinental differentiation between M. rigidula and M. rigiduloides is not as distinct as reported by Small et al. (1990). In the NPGS, 9% of European accessions are M. rigiduloides and 15% of Asian accessions are M. rigidula. The pod morphology of European rigriduloides accessions was essentially identical to Asian accessions of that species. Conversely, Asian M. rigidula accessions had pods that were nearly identical to M. rigiduloides pods. Thus, the key to pod morphology in Small et al. (1990) would not be useful for differentiating these species in Asia.

Sources of Resistance to Anthracnose in the Annual Medicago Core Collection

N. R. O'Neill and G. R. Bauchan
Soybean and Alfalfa Research Laboratory
USDA, Agricultural Research Service
Beltsville, Maryland, 20705, U.S.A.

Recent interest in use of annual species of Medicago in the U.S. prompted an evaluation of the diversity of germplasm that exists and in the development of an annual Medicago core subset. This collection represents the genetic diversity inherent in over 3200 accessions of annual Medicago (1,2). The collection is being evaluated for sources of resistance to major indiginous diseases, both to enhance the utilization of the species as a crop and as a potential source of new or novel resistance genes amenable for incorporation into adapted perennial alfalfa. Most annual medics are diploids and thus genetically simpler to study and manipulate than alfalfa. Anthracnose caused by Colletotrichum trifolii is a major disease in perennial alfalfa grown in North America. Disease control is based principally on the use of resistant varieties (3). Resistance has been associated with the production of several pterocarpan and isoflavan phytoalexins following infection attempts in germplasm with genes for resistance (4,5). Using standardized methods, 201 accessions from 33 species were evaluated in growth chambers for resistance to anthracnose caused by C. trifolii. M. sativa standard check cultivars Arc (resistant) and Saranac (susceptible) were included. Most annual medic species were highly susceptible. Fourteen accessions representing seven species exhibited more than 60% resistance. These were tested for inducible defense mechanisms based on production of phytoalexins. TLC bioassays of seedling extracts revealed the presence of fungitoxic compounds 72 h after inoculation with C. trifolii. Medicarpin and small quantities of sativan were produced in resistant but not susceptible or non-inoculated lines of M. turbinata (PI 566593 from Lebanon). M. muricoleptis (PI 495401 from Italy) produced an unidentified fungitoxic compound in inoculated seedlings. An accession of M. truncatula produced medicarpin and sativan. Medicarpin, sativan, vestitol, and an unidentified compound were produced in small quantites in M. radiata (PI 459146). In all resistant species producing phytoalexins, spores germinated and produced typical appressoria but failed to produce primary or secondary hyphae. The inducible phytoalexin response exhibited in accessions of M. muricoleptis and M. turbinata may account for their resistance to anthracnose and could be exploited as a potential source of new resistance genes. The mechanism of resistance in the remaining twelve lines exhibiting significant anthracnose resistance appears to be unrelated to the inducible production of phytoalexins. Because little resistance to anthracnose was found in the core collection, resistance should be checked and incorporated into species of annual Medicago intended for use in areas where anthracnose is severe.
References

1. Diwan, N., Bauchan, G.R., and McIntosh, M.S. 1994. A core collection for the United States annual Medicago germplasm collection. Crop Sci. 34:279-285.
2. Diwan, N., McIntosh, M.S., and Bauchan, G.R. 1995. Methods of developing a core collection of annual Medicago species. Theor. Appl. Genet. 90:755-761.

3. O'Neill, N.R. 1996. Pathogenic variability and host resistance in the Colletotrichum trifolii/Medicago sativa pathosystem. Plant Dis. 80: 450-457.

4. O'Neill, N.R., and Saunders, J.A. 1994. Compatible and incompatible responses in alfalfa cotyledons to races 1 and 2 of Colletotrichum trifolii. Phytopathology 84:283-287.

5. O'Neill, N.R. 1996. Defense expression in protected tissues of Medicago sativa is enhanced during compatible interactions with Colletotrichum trifolii. Phytopathology 86: in press.

Influence of Light on Resistance to Vascular Wilt Diseases of Alfalfa

B. W. Pennypackerl, M. L. Risius1, J. J Volenec2, and S. M. Cunningham2
Department of Agronomy, Penn State University, and 2Department of Agronomy, Purdue University
University Park Pennsylvania, 16802, and West Lafayette, Indiana 47907, U. S. A.

Fungal-vascular-wilt diseases of alfalfa are managed through genetic resistance. Resistance to Fusarium oxysporum £ sp. medicaginis is controlled by a dominant gene and an additive gene (1). Resistance to Verticillium albo-atrum is polygenic in U. S. commercial alfalfa but dominant in the English variety Maris Kabul (3). Polygenic resistance to V. albo-atrum failed under 40% ambient photosynthetic photon flux density (PPFD), indicating carbon assimilation is critical to the expression of polygenic resistance (2). To determine whether dominant resistance also requires carbon assimilation, we conducted a factorial shading experiment using F. o. medicaginis to invoke dominant resistance and V. albo-atrum to activate polygenic resistance within clones, which avoided the problem of genetic heterogeneity between clones. Four clones, all with dominant resistance to F. o. medicaginis, but representing the spectrum of polygenic response to V. albo-atrum, were used.

Clones were root-soaked for 15 min. in sterile water, V. albo-atrum, or F. o. medicaginis., then grown for 6 weeks before being cut to 4-cm height and used in the experiment; a split-plot, randomized complete block, with 3 replications. The main-plot was PPFD (ambient and 30% of ambient), and sub plots were a 3 x 4 x 2 factorial of pathogen, clone, and sampling time. Photosynthetic Photon Flux Density Destructive sampling occurred in weeks

3 and 5 of the S-week growing cycle. A Fig. 2. Clone x PPFD interaction. Maris Kabul clone having dominant No effect of PPFD on the Maris resistance to V. albo-atrum was Kabul clone with dominant included for verification of the dominant resistance to V. albo-atrum. resistance response. The experiment
was conducted once and is being repeated. A significant pathogen x clone x PPFD interaction was detected in aerial biomass, leaf and stem dry weight, and height (Fig. 1). Polygenic resistance to V. albo-atrum was lost under 30% ambient PPFD, while dominant resistance to F. o. medicaginis was not affected. A significant clone x PPFD interaction (Fig. 2) was detected when two polygenically-resistant clones and the Maris Kabul clone with dominant resistance to V. albo- atrum were analyzed, verifying the PPFD insensitivity of dominant resistance. Root tissue from one polygenically-resistant clone was analyzed for vegetative storage protein (VSP). No significant pathogen x PPFD interactions were detected, indicating VSP was not contributing to the PPFD sensitivity of polygenic resistance. Unlike the expression of polygenic resistance, the expression of dominant resistance does not appear to be contingent on carbon assimilation.
References

1. Hijano, E. H., Barnes, D. K., and Frosheiser, F. I., 1983. Crop Sci. 23:31-34.
2. Pennypacker, B. W., Knievel, D. P., Risius, M. L., and Leath, K. T. 1994. Phytopathology 84: 1350-1358.
3. Viands, D. R. 1985. Crop Sci. 25: 1096-1100.

PERSISTENCE AND LEVELS OF CROWN AND ROOT ROT DISEASES IN ALFALFA
CULTIVARS UNDER MOWING AND ROTATIONAL GRAZING MANAGEMENT
SYSTEMS

A. Navarro and E. H. Hijano
INTA - Argentina

Alfalfa is the primary forage crop in the Pampean Region of Argentina where approximately 5.0 million ha area grown annually. Even though alfalfa cultivars are being tested under cutting and rotational grazing, the effects of both management system have never been compared in simultaneous trials using the same participants, at the same location during a 4-year period.
The objective of these experiments was to compare persistence and levels of root diseases in alfalfa cultivars subjected to rotational grazing and mowing.
The two trials were established in April, 1989 at Manfredi Exp. Stn, Cordoba, Argentina. The grazing trial was a randomized complete block design having 800 m2 plots with 3 replicates. Plots were grazed with Holstein steers using a variable stocking rate according to the available forage in a 7-day rotational system with 35-day resting period.
The same cultivars were planted in 9 m2 plots, using a randomized complete block design with 4 replicates and mowed at the 3rd. or 4th. day of grazing in the "grazing" trial.
After 4 growing seasons, surviving plants were sampled and scored. In the cutting trial, all plants from a 5 m2 cutting area were extracted. The same area was sampled from each grazed plot, digging up plants from five m2 subsamples randomly located within each 800 m2 plot.
Surviving plants were counted and evaluated in a 1 to 5 scale for corky root rot and crown root rot complex severity levels.
Average surviving plants was higher under grazing (14.38 pl/m2) when compared with the cutting (9.62 plt/m2) management system. In both experiments, significant differences in persistence were detected among cultivars. Monarca SP INTA and Victoria SP INTA had the highest values of surviving plants/m2 whereas Cuf 101 and WL 605 had the lowest persistence.
Average severity levels for CRRC and corky root rot were almost identical in both trials being the ranking of cultivars, in terms of disease severity index, quite similar.
From these results, it is concluded that the management system did not affect differentially neither the persistence nor the ranking of alfalfa cultivars measured as levels of crown and root diseases after 4 growing seasons. Cultivar genotype is the main characteristic that determines persistence or crown and root diseases levels, independently of the management system applied for harvesting the forage.

POST-HARVEST FUNGAL RESISTANCE IN ALFALFA: CULTIVAR RESPONSE AND
MECHANISMS

V.I Babij1, K. M. Wittenbergl, and S.R. Smith, Jr 2
Animal Science Department Plant Science Department
University of Manitoba, Winnipeg, Manitoba R3T 2N2

Alfalfa cultivars have been developed for desirable qualities such as feed quality, winter hardiness, grazing tolerance, but no selection protocols have been developed for post harvest fungal invasion. An effective method for screening plants would be beneficial for development of cultivars that have increased resistance to saprophytic organisms during field wilting and storage. Fungal organisms, which include the genera Mucor, Absidia, Rhizopus, Aspergillus and Humicola, develop in inadequately dried hay and reduce hay quality. These organisms also produce spores which are a major cause of respiratory problems for livestock and livestock producers. Preliminary studies have suggested that certain alfalfa genotypes show resistance to mold growth after harvest.
One objective of this study was to verify a leaf screening procedure developed by Wittenberg et al, ( in preparation) for detection of plants resistant or supportive of fungal growth. The screening procedure involves random selection of alfalfa leaves from individual plants, placement of selected leaves onto agar plates, spraying with a fungal spore suspension, incubating at 25 degrees C, and evaluating the fungal biomass coverage on the plated leaf material after the incubation period. Two verification studies were conducted on individual genotypes previously identified as having low, variable and high susceptibility to fungal growth after harvest. The first study simultaneously compared the leaf screening procedure with mold content of incubated stem and leaf material harvested from the entire plant. The second study compared fungal development for these same genotypes during field wilting and bale storage. The results of the first study indicate that the visual evaluations from the screening procedure correspond to the mold content of the whole plant harvested material, however minimal fungal growth across all genotypes in the second study made comparative evaluation impossible.
The second objective was to develop resistant and susceptible populations using the leaf screening procedure. A population of approximately 1000 genotypes representing 22 cultivars was screened for genotypes that showed reduced post-harvest fungal invasion for the resistant population, and the susceptible population included genotypes that supported fungal growth. Studies are being conducted to evaluate these populations for potential physical, chemical and microbial characteristics associated with increased resistance to post harvest fungal growth.
In conclusion, the association between post-harvest fungal resistance of individual leaves in petri plates and mold growth in swaths and bales has yet to be verified. Further research will be conducted this summer in an attempt to relate laboratory and field resistance.
Reference
Wittenberg, K. Smith, R. and Katepa Mupondwa, F. 1995 Screening Methodology for Post-Harvest Fungal Resistance in Alfalfa. In preparation.

Resistance in the Annual Medicago Core Collection to Two Isolates of the Downy Mildew Fungus from Alfalfa

J. R. Yaege and D. L. Stuteville
Department of Plant Pathology, Kansas State University
Manhattan, Kansas, 66506, U.S.A.

Peronospora trifoliorum de Bary causes downy mildew of alfalfa throughout the temperate regions of the world, but little is known about its host range within the annual Medicago species. Using Standard Tests procedures (2), we evaluated the reaction of approximately 50 seedlings each of 199 of the 211 accessions of the annual Medicago Core Collection (1) to P. trifoliorum isolates I7 and I8 from alfalfa from Kansas and southern California respectively. A wide range of symptoms and intensity of P. trifoliorum conidium production occurred. Plants not supporting conidium production were considered resistant and most resistant plants expressed no symptoms or a faint stipple. However, the cotyledons of some resistant plants in a few accessions were killed by a hypersensitive reaction to P. trifoliorum. In 150 accessions, no plants supported conidium production. Species with all plants resistant were M. ciliaris, M. constricta, M. coronata, M. disciformis, M. doliata, M. granadensis, M. intertexta, M. italica, M. laciniata , M. lesinsii , M. littoralis, M. minima, M. muricoleptis, M. orbicularis, M. praecox, M. rigidula, M. soleirolii, M. truncatula, and M. turbinata Accessions with susceptible plants are presented in Table 1. The number of accessions with no susceptible plants for species listed in Table 1 were M. lupulina 5, M. murex 4, M. polymorpha 33, M. rugosa 6, and M. scutellata 13.

References
1. Diwan, N., Bauchan, G. R., and McIntosh, M. S. 1994. A Core Collection for the United States Annual
Medicago Germplasm Collection. Crop Sci. 34:279-285.

2. Stuteville, D. L. 1991. Downy Mildew. D-5. In C. C. Fox, R. Berberet, F. A. Gray, C. R. Grau, D. L. Jessen,
and M. A. Peterson (ed.) Standard Tests to Characterize Resistance in Alfalfa Cultivars. North American
Alfalfa Improvement Conference.

Molecular markers associated with resistance to downy mildew in tetraploid alfalfa
D. E. Obert, D. Z. Skinner, and D. L. Stuteville
Department of Agronomy, USDA-ARS, and Department of Plant Pathology
Kansas State University. Manhattan, KS 66506

The identification of molecular markers associated with disease resistance could provide a screening method for selection of resistant genotypes, and thus prove valuable in the development of improved cultivars. A segregating population derived from crossing two inbred lines is not possible in alfalfa due to its inbreeding depression (Busbice, 1968). Therefore alternative methods of population development for genetic analysis must be employed. Several populations differing only in one trait of interest already exist. Two such populations, UC 123 and UC 143 differ only in that UC 143 was selected from UC 123 on the basis of resistance to isolate I-8 of Peronospora trifoliorum de Bary, causal agent of downy mildew in alfalfa (Lehman et al., 1983). Since the basis of selection between these two populations is resistance to downy mildew, any differences between the two populations, except those occurring within limits of random allelic fluctuation, should be associated with downy mildew resistance. The two populations, each composed of 48 plants, were screened with 41 RAPD primers and 105 polymorphic molecular markers were generated. Marker combinations (2-way and 3-way) were determined and many significant marker and marker combinations were identified.

Analysis to determine genetic distances between individual plants was performed to reduce the population size to be investigated from 96 to 72 in order to perform AFLP analysis on a fewer number of plants. AFLP analysis, although very preliminary, appears promising since each primer combination tested thus rar has generated over 30 molecular markers, of these markers, an average of 12 from each primer combination were polymorphic.

Additional research is underway to confirm the association of the molecular markers with disease resistance by the use of F, and S, progeny.

References
Busbice, T.H. 1968. Effects of inbreeding on fertility in Medicago sativa L. Crop Sci. 8:231-234.

Lehman,W.F.,D.L. Stuteville,M.W.Nielson, and V.L.Marble. 1983. Registration of UC 123 and UC 143 Alfalfa Germplasms. Crop Sci. 23:403.

Dynamics of Apothecial Populations of Sclerotinia trifoliorum

P. Vincelli, J. C. Doney, Jr., and L. Wang.
Department of Plant Pathology
University of Kentucky
Lexington, KY 40546-0091.

Populations of apothecia of Sclerotinia trifoliorum were monitored during the autumn/winter of 1991-92 (one site) and again during 1995-96 (two sites). In all instances, apothecia first emerged between 16 Oct and 25 Oct and were produced over a 9- to 12-wk period. Apothecial density reached a seasonal peak within two weeks of first detection and generally was highest during the first 4-6 wk of production, notwithstanding population fluctuations due to prevailing weather conditions. Using weather data available for the site in 1991-92 and for one site in 1995 96, we detected apothecia after the mean daily soil temperature at a 10-cm depth under grass or alfalfa dropped below 12.8 deg C followed by 0.1 cm or more of rainfall, although stipes probably were developing prior to these conditions. Nighttime low temperatures at or below -5°C were associated with reductions of 50-100% in apothecial density. Studies to understand the dynamics of apothecial populations of S. trifoliorum and the influence of weather events are continuing.

References

G., and Knight, W. E. 1982. Formation of apothecia by sclerotia of Sclerotinia trifoliorum and infection of crimson clover in the field. Plant Dis. 1021-1023.

Valleau, W. D., Fergus, E. N., and Henson, L. 1933. Resistance of red clovers to Sclerotinia trifoliorum Erik., and infection studies. Bull. Kentucky Agric. Exper. Stn., Bulletin No. 341, pp 116-131.

Williams, G. H., and Western, J. H. 1965. The biology of Sclerotinia trifoliorum Erikss. and other species of sclerotium forming fungi: Apothecium formation from sclerotia. Ann. Appl. Biol. 56:253-260.

Disease Resistance of Australian Lucerne Cultivars

G.C.Auricht
South Australian Research and Development Institute.
GP0 Box 397 Adelaide SA 5001, Australia

Lucerne cultivars commercially available in Australia were inoculated with one of six different pests or diseases in separate glasshouse experiments for detailed comparison and testing of a new design. Seed of 28 cultivars and one check was sown in single rows in 30 x 40 cm plastic boxes. Seven lines and one control were sown per box with three replicates giving a total of 12 boxes per experiment. Seedlings in each row were thinned to 20 (aphid tests) or 40 (fungal diseases) then inoculated with either Blue Green Aphid (BGA), Acrythosiphon kondoi; Spotted Alfalfa Aphid (SAA), Therioaphis trifolii f. maculata; Anthracnose, Colletotrichum trifolii; Spring Black Stem and Leaf Spot, Phoma medicaginis, Staganospora, Staganospora meliloti or Phytopthora Root Rot (PRR), Phytopthera medicaginis. Experiments were conducted and scored using techniques based on the "Standard Tests to Characterize Alfalfa Cultivars". Trials were analysed using percentage of resistant plants except in the case of Phoma where the mean damage score (1 to 5) for plants in each row was analysed and BGA where both were analysed. For SAA an analysis of variance was conducted with control rows and box effects removed after check plants died. A square root transformation was required before analysis of the Phoma and BGA results. BGA %Res is high with inclusion of 3's.

Inclusion of checks increases precision, allows for box effects and enables valid comparisons between experiments. BGA mean score gives a better comparison than "%Res".

Alfalfa Germplasm Resources and Evaluation of their
Disease Resistance in Semi-arid Areas of Gansu Province,
People's Republic of China

Ma Zhengyu

(Lanzhou Institute of Animal Science, Chinese Academy
of Agricultural Sciences, Lanzhou, Gansu
730050, P.R. China)

General Situation. After more than ten years elapsed, we have collected, planted for observation, and evaluated 293 copies of various forage plant species at fixed positions of several semi-arid areas, mountainous and flat plain, in Yuzhong County, Central Gansu, near the Provincial Capital, Lenzhou. Among them, there are 256 copies of legumes (occupied mainly at a percentage of 87.37%), and the focal point of collection was alfalfa germplasm resources, 163 copies in all (63.67% in legumes and 55.63% in all respectively). As for germplasmic studies, with the exception of phenological stages, agronomic characters, purity, etc., the main work of disease resistance evaluation was done for detection of such pathological pests as, alfalfa downy mildew, brown spot and brown leafspot diseases, and so on. Alfalfa disease plots and germplasm propagation garden were also set up on trial. II. Evaluation of Anti-Downy Mildew Genetic Resources in Alfalfa.
i. Method of Study. Comparative studies on manipulative inoculation in case and meantimely direct inoculation on land test were done for pathological description. The 5-grade statistical method was used to inspect disease infection as usual. 2. Test Results. 68 copies of field and 36 copies of indoor inoculation cases were tested, the results showed that the infection rate indoor reacted fluctuately with the lowest percentage of 4.8% and highest of 81% in accordance with their respective grades affected. The alfalfa cultivars tested with rate of infection under 20% were as follows: Saramas, No. 80-19 (introduced cv.), U.S. No. 2, Zhaodong (local land race), Netherlands Lucerne, Resis, Oakland and so on; those between 20% and 50%, Xinjiang Anti-drought, Rhizoma, Vernal, Zhaodong No. 18, Northwest A, No. 79-76 (introduced cv.), Humeng No. 18 (Local land race) and so on; as well as those higher than 50%, only 3 in number, Tumu No. 2, French lucerne and a local land race, Tianshui, with respective rates of 55.5-68.2%, 57.5-68.2% and 78.0-81.0% (indoor records). Contrarily, those in showing the natural rate of infection in field higher than 50%-81% were only two, namely, they are Qingyang (Gansu land race) and Hetian (Xinjiang land race). 3. Final explanation for evaluation results. Our discussion will center on disease resistance against alfalfa downy mildew (Peronospora aestivalis), the preliminary viewpoint is that almost and even always, the most resistant cultivars and land races in alfalfa are those introduced from abroad, such as U.S. No. 2, 81 85, 80-19, 81-125, Netherlands Lucerne, and so on that belonging to the typically high-resistant and resistant model plants, but all those local land races planted in China are susceptible. Among the 30 alfalfa land races tested, we just have only three which are said to be highly resistant to downy mildew, and the remained 27 are nothing but merely included to be medium susceptible or highly susceptible. III. Sieved Type Breeding of Alfalfa Genetic Resources.
1. Sieved Type Breeding Test in Selecting Anti-Downy Mildew Cultivars and Land Races of Alfalfa. In the midst of 148 copies of alfalfa cultivars and land races introduced from 14 countries abroad, we have done lots of work on manipulative inoculation, repeated plus reversal selection and some other methods, still up to the present, we just tentatively selected 18 out of those alfalfa cultivars and land races forementioned to be the most productive and highest downy mildew resistant category. Acknowledgment
We heartily thank Professor Wu Renrun's Kind recommendation and disinterested assistance to help consulate the paper to its perfection extent in the presenting to the 35th NAAIC poster session.

Detection of Antixenosis Resistance Mechanism to Blue Aphid
in an Argentine Alfalfa Population
J. O. Gieco, D. H. Basigalup and E. H. Hijano
EEA Manfredi, I.N.T.A., Cordoba, Argentina

Genetic variability in alfalfa regarding reaction to blue alfalfa aphid (BAA), Acyrthosiphon kondoi Shinji, has been found; however, there are practically no studies on identification of resistance mechanisms.
In a previous paper (1), we reported on the detection of antibiosis and tolerance mechanisms in 36 alfalfa clones ranging from extreme susceptibility to high resistance to BAA. Four clones were identified as primarily antibiotic, thirteen clones were identified as primarily tolerant, and two clones simultaneously exhibited both.
The same plant population was also tested for the presence of antixenosis, a resistance mechanism defined as plant unattractiveness for insect food, shelter, or oviposition. In this paper we are going to present the results of this test.
The 36 alfalfa plants were selected after a severe BAA attack at the Manfredi Exp. Stn.-INTA breeding nursery in 1991. All plants were cloned on vermiculite in the greenhouse and, after 45 days, transplanted to plastic squared containers (33 x 33 x 18 cm) filled with a 3:1 sterilized organic soil:perlite mixture and surrounded with a transparent PVC film covered with nylon fabric. Aphid population samples were taken from the field, purified, and increased in the greenhouse. The test was performed under 18 24 °C, 80% RH, and 15 hr daylength conditions.
An incomplete randomized block design with 6 replications was used. Each container received 9 different clones in a circular disposition, according to a target-like diagram. In the center of each target, 50 adult blue aphids were liberated; after 12, 24, and 48 hr the number of aphids on each plant was recorded. Standardized data were used to perform a cluster analysis (average linkage method) to group clones according to insect preference based on pseudo F and pseudo t2 values.

Eight clusters were formed. Average aphid numbers from the most antixenotic to the most susceptible clusters were 2.28, 2.43, 3.28, 11.50, 10.92, and 10.09 at 12, 24, and 48 hr after liberation, respectively. The most antixenotic group included clones 2, 8, 10, 11, 12, 15, 16, 24, and 32. While clone 32 had been previously identified as antibiotic (1), in this test was also defined as antixenotic; this is in agreement with the idea that a plant can express more than one resistance mechanism.

REFERENCE
(1) GIECO, J. O., E. H. HIJANO and D. H. BASICALUP. 1994. Report 34th NAAIC, Guelph, Ontario, Canada, July 10-14, p. 112.

Are Chemical Residues Sufficient in Killing or Deterring Blister Beetle
Swarms in Alfalfa?

P. G. Mulder, Jr. and K. T. Shelton
Department of Entomology, Oklahoma State University
Stillwater, OK 74078-3033

Cantharidin, the blistering agent produced by blister beetles, was first isolated in 1810 from Lytta vesicatoria (Linn.). It is highly toxic if taken internally. Horses are extremely sensitive to cantharidin. Approximately lmg/kg is the minimum lethal dose for horses (Ray et al. 1980). In Oklahoma, blister beetles complete one generation per year. Although several species are common throughout the United States, the striped blister beetle, Epicauta occidentalis (Werner) is the gregarious insect most commonly associated with alfalfa in Oklahoma. Several authors have suggested means of combating blister beetle problems (Mulder et al. 1993); however, because of the elusive nature of the insect and unpredictability of its peak occurrence, producers either ignore the problem or treat prophylactically just before cutting. Unfortunately, insecticides currently labeled for control of blister beetles in alfalfa have preharvest intervals of seven days or more. Concerned over the immigration of swarms from adjacent areas into treated fields, producers often choose to ignore waiting periods and harvest before preharvest intervals have expired.
To further elucidate the subject of residual control of Sevin insecticide on blister beetles, a greenhouse study was conducted using nearly 3,000 beetles captured in the field. Control of blister beetles, ranging from 92% to 53.4% was obtained from 1 to 120 hours after treatment with Sevin XLR. Over the first five days, percent mortality decreased by 10.6% in treated plants. In untreated plants, percent mortality decreased by 28%. Higher rates of mortality in untreated plants was due, in large part, to mortality caused by manual infestation. Based on analysis of forage consumed during the trial, data suggest that beetles are not only effectively killed by chemical residues but are also deterred from consumption of foliage.
Information from this study and two extension publications was presented at approximately six regional meetings across Oklahoma in 1995. Over 500 producers had an opportunity to ask questions concerning their alfalfa harvest management practices.
References

- Mulder, P. G., Shawley, R. V., and Caddel, J. L. 1993. Blister beetles and alfalfa. Oklahoma State University, Oklahoma Cooperative Extension Service Fact Sheet No. F-2072.

- Ray, A. C., Post, L. O., Hurst, J. M., Edwards, W. C. and Reagor, J. C. 1980. Evaluation of an analytical method for the diagnosis of cantharidin toxicosis due to ingestion of blister beetles (Epicauta lemniscata) by horses and sheep. Am. J. Vet. Res. 41: 932-933.