Plantain and banana, both Musa spp. hybrids, constitute a major component of domestic food production in Haiti. In response to grower reports (Faculty of Agronomy and Veterinary Medicine, Port-au-Prince, personal communication) of up to 60% incidence of “toppling disease”, we conducted a preliminary study in July 2018 at four plantain farms in Arcahaie, Haiti. Symptoms observed on up to 30% of plants (n = ∼100) included a brown necrotic soft rot lesion on pseudostems extending up to 1 m in length and several centimeters deep. Toppling occurred at fruit onset, with plants folding over at the point of the soft rot. Six plants representing the four farms were sampled: 6-cm² sections were excised from the lesion edge, surface sterilized with 0.6% sodium hypochlorite, rinsed, and macerated with sterile tap water, and then the suspension was streaked onto nutrient agar (NA) and Miller-Schroth medium agar (MSM) plates, yielding six bacterial isolates. Colonies were round, glossy, convex, translucent gray-white on NA, produced a transient red-orange color change on MSM, and were aerobic, gram-negative, oxidase-negative and hypersensitive response negative on ‘Bonnie Best’ tomato and ‘Hicks’ tobacco (Schaad et al. 2001); all the above tests were repeated twice. All six isolates were identical in initial 16S rDNA sequencing, so pathogenicity of two randomly chosen isolates (G18-1365 and G18-1376) was tested on disease‐free ‘Dwarf Cavendish’ banana plantlets by drilling a 2.5-cm hole into the pseudostem, inserting sterile cotton-tipped swabs drawn through 48-h cultures on NA (negative control was dipped in sterile water) into the drilled incision, and sealing with Parafilm. Each plant was wounded once, with five plants per isolate or control. Rotting symptoms consistent with the field observations were observed 4 weeks after inoculation for all isolate inoculations; control plants remained symptomless. Reisolation and identification to fulfill Koch’s postulates were completed with results identical to the first isolations and tests. Genomic DNA was boil-extracted as previously described (Dashti et al. 2009) from the two isolates grown 24 h on NA. The amplification and sequencing of 16S rDNA, rpoB, phoE, and infB gene targets were carried out using primers and protocols previously described (Rosenblueth et al. 2004). One sequence for each isolate’s gene target (eight total) was deposited in NCBI GenBank (accessions MK217521 to MK217522; MK235993 to MK235998). In BLAST searches, sequence fragments of 16S rDNA, rpoB, phoE, and infB from isolates G18-1365 and G18-1376 resulted in 99% identity with Klebsiella variicola strains. Multilocus sequence analysis (MEGA X, Kumar et al. 2018) of those sequences yielded a concatenated phylogenetic tree that placed both isolates within the K. variicola clade. Isolates G18-1365 and G18-1376 have been deposited at the BCCM, Ghent University (accessions LMG 31122 and LMG 31123, respectively). K. variicola has been reported to cause similar bacterial rots in banana and carrot (Chandrashekar et al. 2018; Fan et al. 2016). Before management can be prescribed, further studies are necessary to determine disease incidence and if this species is the sole cause of toppling or is part of a larger disease complex, and to understand the mode of transmission within and among plantain farms. Additionally, similarity to human pathogenic strains should be studied to determine if there is a human health risk posed by K. variicola isolates causing soft rot in plantain.