Balloon vine, Cardiospermum grandiflorum Sz. (Sapindaceae) is a perennial, slightly woody climber which occurs naturally in tropical and subtropical Africa, Asia, Australia and America. Balloon vine is highly invasive along forest margins, watercourses and urban open spaces in subtropical regions. The magnitude of invasion by balloon vine in recent years has been quite severe, threatening to biodiversity along the coastal regions of KwaZulu-Natal.

​

​​​

​​

​​​

​

​​

Mechanical control of balloon vine is extremely difficult and costly, as dead plant material has to be removed to restore exposure of the understorey to sunlight. Chemical control is similarly problematic because of non-target damage to underlying vegetation. Biological control is thus considered the only feasible option to curb the invasion of this vine in South Africa.

A biological control programme​ for balloon vine was initiated in 2003​. However, the programme became complicated due to uncertainty regarding the origin of two closely related plant species (i.e., Cardiospermum corindum and Cardiospermum halicacabum), which are cosmopolitan from the central to southern parts of America and Africa. Through phylogenetic studies, it has since been determined that C. corindum and C. halicacabum are of Southern African and South American origins, respectively. The African native range of C. corindum implies that biological control of C. grandiflorum in South Africa should be limited to agents that develop exclusively on this weed.

Field host range and open-field studies in the native range have so far revealed that the seed-feeding weevil Cissoanthonomus tuberculipennis​ (Coleoptera: Curculionidae) and the flower gall-forming midge Contarinia sp. (Diptera: Cecidomyiidae) are the only agents that are restricted to balloon vine. The weevil C. tuberculipennis has since been approved for release in South Africa, while attempts are being made to rear the midge Contarinia sp. in quarantine laboratory. Seventeen months after its release, C. tuberculipennis had fully established at 12 of the 13 sites, and has dispersed for over 50km, destroying 13 to 33% of the seeds at release sites in KZN, Limpopo, Mpumalanga and Gauteng provinces of South Africa. Once the spread of the weevil has saturated throughout the geographic distribution range of balloon vine, it is expected that its population density and infestation levels will increase. 

Whilst the seed weevil C. tuberculipennis is widely expected to curb the spread of balloon vine, the current infestations of the weed will remain unless they are physically removed. Therefore, there is still a need to complement this agent in order to enhance biological control of balloon vine in South Africa. Future explorations will therefore include search for stem-attacking natural enemies as these have not been thoroughly explored during the past 10 years in the native range.