A very vigorous and productive coffee plant found in a plantation in the State of São Paulo was studied and found to be a tetraploid (2n=44). In this respect it appeared to be similar to the species C. arabica but in its self-sterility it seemed to be more like the diploid types (2n=22) of coffee. It was at first thought that this plant could be a tetraploid form of a diploid species such as C. liberica, or C. Dewevrei. The study presented herein, however, reaches the conclusion that this coffee plant is more likely to be the result of hybridization between C. arabica and C. Dewevrei. In a study of microsporogenesis it was observed that uni - and bivalents and less frequently tri - and tetravalent chromosomes were formed. If one assumes that the somatic complement of this plant is made up of the 22 haploid arabica chromosomes plus the 22 diploid Dewevrei chromosomes, then it can be estimated that the arábica chromosomes will form 0 to 6 bivalents as is known from the study of a haploid plant of C. arabica ; and if the Dewevrei chromosomes form 11 bivalents, then in the micros-porocytes of the hybrid one would expect to find 11 to 17 bivalents ; in the present study we found most frequently 12 to 16 bivalents which is within these limits. This, then, supports the hypothesis that the plant studied is a hybrid between C. arabica and C. Dewevrei. Study has shown that anaphasic distribution of the chromosomes is irregular and that pollen grains are formed with n = 15 to 28 chromosomes; most frequently (ca. 80%) the grains have n = 20 to 24. Controlled hybrids were made using this plant as the mother tree and C. arabica (2n=44), C. Dewevrei (2n = 22) and C. Dewevrei (2n = 44), as sources of pollen. A study of the chromosome numbers in these hybrids showed that fertile eggs were formed with n = 20 to 24 chromosomes. In the plant studied approximately 31% of the pollen grains were observed to be sterile. It was also observed that some ovules do not develop and one frequently encounters empty or only-one-seeded fruits. It is suggested that the sterility of the ovule and pollen grains may be due to the same factors.