52 SEEDS AND PLANTS IMPORTED. 



45540 to 45553— Continued. 



seen it cultivaled in Antigua and am tokl that it occurs wild in that 

 region as well. 



"The wild trees, which grow on rocljy, rather dry slopes, reach 20 

 feet in height. In April and May they produce numerous flowers 2 

 inches in diameter, white upon first opening, but later becoming bright 

 pink. When in full bloom the tree is very decorative in appearance and 

 worthy of a trial in the warmest sections of the United States." 



45551. (Undetermined.) 



"(No. 208a. November 8. 1917.) A flowering vine from the summit 

 of the Cachil Mountains, north of Salama. Ba.la Vera Paz; altitude 

 5,250 feet. 



" This plant is occasionally seen climbing over shrubs and small trees. 

 It does not make very luxuriant growth, but produces clusters of small 

 red flowers which are very attractive. The flowers are followed by 

 winged seed capsules. For trial in California and Florida." 



45552. Gliricidia meistophylla (Donn. Sm.) Pittier. P^abacese. 



"(No. 209a. November 8, 1917.) Seeds of a leguminous shrub from 

 the mountains of northern Baja Vera Paz." 



45553. Persea Americana Mill. Lauracese. Avocado. 



(P. gratissima Gaertn. f.) 



"Avocado seeds to be grown for stocks." 



45554 to 45557. 



From Buitenzorg, Java. Seeds presented by the director of the Botanic 

 Gardens. Received November 30, 1917. 



45554. Pavetta zimmermanniana Valet. Rubiacese. 



A small rubiaceous tree or shrub, with opposite, nearly elliptic leaves 

 and clusters of small, slender-tubed white flowers. 



" The remarkable researches of Zimmerniann and Faber detailed in the 

 Jahrbiicher fiir AVissens<:-haftliche Botanik, vol. 51, p. 285. 1912, and 

 vol. 54, p. 243, 1914. make this species of unusual interest. Faber has 

 proved that the leaves of this and of several other species of Pavetta, 

 Psychotria, and possibly other genera of the Rubiacese contain colonies 

 of a nonmotile. nitrogen-fixing bacteriinn which he names Myco-bacterlum 

 ruhiaceanui). The bacteria of this species almost invariably inhabit the 

 micropyle of the young seed and, when the seed germinates, grow through 

 certain stomata of the very young leaves and into the intracellular spaces 

 formed in the leaf tissues around these stomata. Cavities are formed 

 through the growth of the epidermal cells, which later close entirely and 

 make bacterial nodules wh'ch are deeply embedded in the leaf tissues. A 

 single leaf may have several dozen of these symbiotic bacterial nodules. 

 Faber was able, by treating the seeds with hot water and a sublimate 

 solution, to kill the inluibiting niyco-bacteria and, later, to infect part of 

 the seedlings grown from these seeds with pure cultures of the Iiacterium. 

 The artificially infected seedlings grown in soil free from combined nitro- 

 gen grew well and remained healthy for four months, whereas those not 

 so infected turned yellowish white and died in three or four weeks. The 

 plants from unsterilized seeds produced leaves bearing many more bacterial 

 nodules than did those from sterilized seeds which were later artificially 

 inoculated. In view of the facts that these rubiaceous plants with 



I 



