03782nam a2200313 a 450000100080000000500110000800800410001910000170006024500940007726000400017130000100021150000230022152028980024465000270314265000240316965000270319365000150322065000250323565000260326065000240328665000310331065000220334165000120336365300200337565300170339565300220341265300150343465300190344915602592001-10-18       1978    bl uuuu  m   00u1 u    #d1 aGOMES, D. T.  aEstablishment methods and comparative persistence of five tropical legumes in grass sods.  a[S.l.]: University of Floridac1978  a143p.  aPhD. Dissertation.  aThe main purpose of the research was to determine the effect of several seedbed-preparation methods on the establishment and persistence of five tropical legumes in two tropical grass sods. As an additional objective, the effect of the legume inclusion in the grass sods on the quantity and quality of the forage produced by the mixtures was also studied. Two small-plot *field experiments were conducted at the Agricultural Research Center, Fort Pierce, Florida, from 1 july 1976 to 30 Sept. 1978. The two experimental areas used were 3-year-old stands of 'Pangola' digitgrass (Digitaria decumbens Stent.) and 'Bigalta' limpogross (Hemarthria altissima (Pior.) Stapf et C.E. Hubbard). Five seedbed preparation methods were compared: no tillage, light discing, sod-seeding, and complete seedbed preparation. Five legumes were tested: 'Siratro' (Macroptilium atropurpureum (DC.) Urb.), centro (Centrosema pubescens Benth.); joint vetch (Aeschynommene americana L. ), 'Verano' Caribbean stylo (Stylosanthes hamata Taub.), and 'Florida' carpon desmodium (Desmodium heterocarpon (L.) DC.). The treatments were arranged in a split design with five replications. Even though all tested established quite well in mixture with pangola digitgrass, Siratro and joint vetch were the only ones that persisted and actually made a substantial contribution to the total dry-matter yields of the pangola-legume mixtures during the growing seasons ot 1977 and 1978. A pangola-legume minture containing either 50% Siratro or 50% joint vetch provided twice the dry-matter, and four to five times the crude protein yields of pangola digitgrass grown alone. In the Bigalta limpograss experiment, as in pangola digitgrass, all legumes established quite well. joint vetch, carpon desmodium, and centro performed better than the other legumes. The persitence of these legumes was most likely due to their better adaptation to the poorly drained soil conditions of the experimental area. The legume contribution to total dry-matter yields of Bigalta-legume mixtures in the 1977 and 1978 growing seasons was not very marked. The highly competitive growth chacteristic of Bigalta is believed to be the main reason for the low legume contribution to uields from the mixtures. Results of the experiments suggested that seedbed treatments might not be as critical for persistence as it is for establishment of the legumes. In the year of establishment, all legumes were significantly affected by the type of seedbed preparation and there was a substantially higher number of plants in some treatments than in others. These differences, however, were not evident 1b and 2 years after establishment when all legumes tended to reach a rather uniform population in all seedbed-preparation treatments. Joint vetch was the only legumes for which seedbed treatments were still execising their influence in the 1978 growing season.  aDesmodium heterocarpon  aStylosanthes hamata  aAeschynomene Americana  aCarrapicho  aCentrosema Pubescens  aLeguminosa Forrageira  aLeguminosa Tropical  aMacroptilium Atropurpureum  aPlanta Forrageira  aSiratro  aEstabelecimento  aFeed legumes  aPlanta leguminosa  aStylosante  aTropical crops