This is a morphologically variable species that is widespread within Cuba. It was circumscribed broadly by McDowell (1995), based on extensive field work, but was narrowly parsed into several species and infraspecific taxa by Borhidi (1989, 2002) and Borhidi et al. (2018). Exostema shaferi was treated by Borhidi (2002) as a synonym oif Solenanddra myrtifolia, but is treated here as a synonym of Solenandra ixoroides following McDowell (1995)
The type of Solenandra ixoroides was cited by Borhidi et al. (2018) as Hooker's plate in the protologue, but the place where this plate was chosen as the lectotype has not been found and McDowell (1995) cited an original specimen that was physically seen at K as the holotype. If there is some doubt about the uniqueness of Hooker's specimen as the holotype of the name, the specimen still has priority over the plate as a lectotype.
Solenandra ixoroides was treated by Borhidi et al. (2017) under the resurrected, supposedly older name Exostema valenzuelae, but that name was never validly or legitimately published. This name was not validly published in its original use by A. Richard in his 1850 Rubiaceae of Cuba, but its status there has been widely confused . This name was given on his plate there, and in the same work was mentioned in a note under Exostema parviflorum. The note said, however, that the "bad" name Exostema valenzuelae had been used incorrectly for this species on the illustration, and that Richard did not actually accept this name in this work. This statement that the name was not accepted by its author makes it invalid. In spite of this, the name Exostema valenzuelae was catalogued as a name published by Richard by Index Kewensis, perhaps based on the figure without cross-reference to the text, and mistakenly taken up by others. (Richard's figure alone could be adequate to validate the publication of this name if he had not explicitly invalidated it in the same work.) In 1928 Urban discovered Richard's "bad" name, again perhaps on the figure, and he adopted it and cited it as the oldest and correct name for this species. However, the valid publication of this name dates only from Urban's article, and Urban cited two other, older, validly published names in synonymy there with Exostema valenzuelae. Thus, Urban did finally publish this as as a valid name but it was a superflous and illegitimate name. This problem was not noted by Borhidi (1989), who adopted this illegitimate name for plants of Cuba and named several infraspecific taxa. Later (Borhidi, 2002) did note this problem, and stated that Solenandra ixoroides was the oldest legitimate name for this species and synonymized Exostema valenzuelae with it. Later, however, Borhidi (2017) inexplicably again adopted the illegitimate name Exostema valenzuelae (as Solenandra valenzuelae) for these plants. The infraspecific taxa he described in Exostema valenzuelae are valid names, but because they were described in an illegitimate species name, they are not correct and must be transferred to a legitimate species name for use.