Would the real Inga edulis please stand up
Ice cream bean is a South American fruit tree sometimes seen for sale in Nurseries or growing in Perth and surrounding areas, but it's not common in WA and there's limited local knowledge regarding the species. No real development work has been done on them globally, so there are no named cultivars, and if you manage to source one here it will be a seedling, inevitably labelled I. edulis.
The species name comes from the Latin word for 'edible' and the plant you acquire may ultimately produce edible fruit. The problem is that the genus consists of around 300 species and about 50 of these produce edible fruit. The following pictures (taken from Amazon River Fruits: Flavours for Conservation, Professor N. Smith, R. Vasquez & W.H. Wust (2007), Missouri Botanical Gardens Press, 272pp), illustrate just some of the variation in fruit forms of different edible species. See also a review of this book: the link is in the Essays list. The genus is divided into 14 sections with I. edulis in section 12. Most of those shown in the pictures are in other sections, but as might be guessed from the similar pod morphology, I. ingoides is also in section 12. So with all these other possible confounding species that could have slipped into the original sourcing, propagation and marketing chain somewhere, is it possible to firm up your confidence that your new plant is what you expect it to be?
With this level of diversity in the genus, the answer to being able to nail it down precisely to edulis is "no", short of using a full taxonomic key. But you can fairly easily exclude many possibilities with a few simple checks:
- Early ones while still in the non-fruiting juvenile period are that the plant should be pubescent, young shoots should be slightly angular and leaves paripinnate with 4-6 pairs of leaflets and a winged rachis between leaflets. Look for a small sessile transversely-compressed (kidney-shaped) nectary on the upper surface of the leaf rachis between each pair of leaflets. The terminal pair of leaflets should be larger than the base pair.
- In the flowering stage the axillary inflorescences should be congested spikes. Flowers should have a tubular calyx and corolla, with 55-100 stamens (1.5-3cm long) and white filaments tipped with tiny pale yellow anthers.
- When it gets into the mature fruit-producing stage, the pods should be roughly cylindrical and longitudinally ribbed, olive-green, indehiscent, and very long (up to 1m or more) and skinny. In many of the other species the pods bulge noticeably around each seed whereas in edulis this is much less pronounced In cross-section, the pod should be shaped like a toothed cog.
- and when you open the pods, the seeds should be elongated (2-3cm long X 1-1.5cm wide), longitudinally-oriented, not flattened, and preferably black.
This pod is Inga edulis
Correct identification is not purely academic as many other species don’t taste quite as nice. Best to be a little cautious when you buy a plant from retail outlets. Often in these matters of possible confusion, you can cut through the dilemma of whether you're spending all your management efforts over some years on raising the real McCoy is to taste a fruit from a local tree satisfying at least the above criteria, obtain scion material and graft onto a seedling. Alternatively you can germinate seeds from such a plant but you may have to wait a little longer for fruit, they may not be exactly true to type and the tree will want to be bigger. The final picture is from a tree growing here in the Perth suburbs showing a pod that clearly is not I. edulis
.
Barry Madsen
