Folia Microbiol. 42 (6), 601-606 (1997) Roystonea regia, a Monocotyledonous Tree, Bears Rhizobial Root Nodules P.S. BASU,A.C. GHOSH and T.K. DANGAR* Department of Botany, Burdwan University, Golapba~ Burdwan 713 104, West Benga~ India Received February 11, 1997 Revised version June 11, 1997 ABSTRACT. The first instance of a monocotyledonous tree Roystonea regia (royal palm) bearing root nodules is reported. They are formed by a Gram-negative bacterium which was tentatively identified as Rhizobiura sp. The nodules are formed on younger roots of the tree mostly in the rainy season. Younger plants also become nodulated. Mature nodules are oval to cylin- drical in shape, unbranched and loosely attached to the roots which show ability to fix nitrogen. Leguminous plants are known to bear nodules on their roots. Those dicotyledonous plants are now divided into three families. The nodules are formed generally on younger roots due to infection with Rhizobium or Bradyrhizobium species which are Gram-negative bacteria (Jordan 1984). These bacteria form root nodules almost entirely on leguminous plants. A strain of Bradyrhizobium has been reported from three species of Parasponia (Trerna), members of Ulmaceae, a family of dicotyledonous plants (Jordan 1984; Trinick 1973) and there are unconfirmed reports of their occurrence in three gen- era of Zygophyllaceae, another family of dicotyledonous plants (Smith and Douglas 1987). Some acti- nomycetes (bacteria) are also known to form nodules on roots of dicotyledonous plants (Smith and Douglas 1987). Monocotyledonous plants are not known to bear such root nodules. This is to report, for the first time, that a monocotyledonous tree Roystonea regia (royal palm) bears root nodules formed by a Gram-negative bacterium which was tentatively identified to be a Rhizobium species. MATERIALS AND METHODS The symbiont was isolated aseptically from healthy root nodules of R. regia and grown in pure culture (Sinha and Basu 1981). This was done after repeated purification through serial dilution and plating. The basal medium for growth was the yeast-extract mineral medium of Skerman with 1% mannitol (YEM) and 0.01% CaCI2"H20 at pH 7.0. The bacteroids were isolated from the nodules following Rigaud and Puppo (1975). The nodules were crushed aseptically and diluted with sterile water. After centrifugation at 300 g, the supernatant was further centrifuged at 7000 g to obtain the pellet of bacteroids. For scanning electron microscopy the bacteria were fixed with 2 % glutaraldehyde and gradu- ally dehydrated passing through different concentrations of ethanol up to 100 %. It was washed with acetone to make it ready for grid preparation. For transmission electron microscopy a drop of bacterial suspension (about 50 cells) was dried over carbon-coated grids, negatively stained with 2 % aqueous phosphotungstic acid (pH 7.0) and observed. Nodule tissues were fixed in 2.5 % glutaraldehyde (0.1 mol/L phosphate buffer, pH 7.4), post- fixed with 1.0 % osmium tetroxide (prepared in the above buffer), stained in 0.5 % aqueous uranyl acetate, dehydrated in graded ethanol-acetone series and embeded in spurr-resin. Ultrathin sections (60 rim) stained with 2 % uranyl acetate and Reynold's lead citrate were observed in a transmission electron microscope. Nitrogenase activity was estimated by the acetylene reduction assay method using gas chro- matography (Postgate 1972). The nodules were placed in a 10-mL vial and capped with a serum cap. One mL of air from the vial was withdrawn with a syringe and immediately filled with 1 mL dry and freshly prepared (from CaC2) acetylene. The air-to-acetylene ratio in the vial was 9:1. Freshly prepared acetylene from CaC2 was used because it was cleaner than technical acetylene (Burris 1974). The con- trol vial containing the nodules was not incubated with acetylene. After 1 h, ethylene produced from *Present address: Division of Entomology, Central Rice Research Institute, Cuttack 753 006, India.