I fairly recently came across a note about an observation made by Justin Thomas at the Institute for Botanical Training that a potential definitive differentiator between I. capensis and pallida (common jewelweed and pale jewelweed) may be the number of teeth on larger leaves, with capensis having 9 or fewer, and pallida having many more, up to 14. I checked all of my observations, and a random selection of several dozen of each species on iNat and it seems to hold true. Exciting, if it does prove correct!
In any case, I put together a table that shows many characteristics of the 2 species, marking the ones that are at least suggestive, up through definitive. It only has the 2 species, but (selfishly) those are the only ones around me so they are the most relevant (to me). I have posted the table as a document on OneDrive as the iNat journal feature does not appear to allow tables, and that is really the only effective way to display the information.
I welcome comments, updates, etc., and most importantly - does this match your experience? I have posted a short summary in the journal area , so comments can go there or here.
It may be worth adding, for readers not terribly familiar with the species, why this discussion is useful- it’s easy to get the idea that one has orange flowers, the other has yellow flowers, easy peasy.
I personally would mainly find this useful for identifying vegetative plants, as both species are common here, often growing together, and pretty consistent and well differentiated in flower color. I notice the chart omits color- is flower color less reliable in other areas, such as where you are in MI?
The note on possible limestone affinity is interesting- if I think about it, the habitats where I find Pallida abundantly all have either high-pH glacial gravel deposits, or calcareous shale exposures (rarely much limestone, except as occasional caprock layers in gorge cliffs) and track well with other calciphilic species.
I don’t know if this helps the discussion but some time ago I had an observation of an I. capensis in Manitoba with almost no red spots. I did a little research and found that spotless I. capensis do occur and they can be confused with I. pallida. The article said one distinguishing feature is that I. pallida have perpendicular nectar spurs and I. Capensis have a curved spur. Unfortunately I didn’t bookmark the source.
Sorry. I didn’t realize the shape of the rear of the flower was already covered in the table. When I first opened the table I thought the table was only one page.
Both Scoggan’s Flora of Manitoba and Budds Flora of the Prairie Provinces resist the idea of I. pallida occuring in Manitoba. Scoggan goes so far as to suggest that all I. pallida specimens collected in Manitoba that he examined were the actually the spotless form of capensis which he states is “common throughout the Manitoba area.” VasCan follows this line as well by excluding pallida from Manitoba. http://data.canadensys.net/vascan/taxon/3659
this might provide you with a control group of observations identified as impatiens capensis or pallida that might all be capensis regardless of flower color - that is if you agree with them :)