Not counting the unusual green violet (Hybanthus concolor), there are 78 documented taxa in the violet family (Violaceae) in New York. This number includes hybrids (of which there are many) and some lower taxa. If we only count violet species, NY has a total of 29. While many species are common, there are some that are quite rare and a few are thought to be extirpated.
Violets can be divided (artificially perhaps) into those that have leafy stems (leaves and flowers arising from an above-ground stem) and those that don’t (leaves and flowers arise directly from an underground stem or rhizome). The latter group is easily identified as the flowers and leaves appear to arise directly from ground level (unless the rhizome happens to be exposed but it is usually at least partially buried).
The stemmed violets are the smallest group, represented by 9 species, of which only 7 are native. The pansies, V. arvensis and V. tricolor, are the non-natives and are distinguished by the presence of large, leafy stipules and petals that form a flat face. We do have one native that fits into the “pansy” group and that is V. bicolor, which is quite rare and restricted to extreme southern NY and a few spots in western NY. These three are also the only annual species that occur in North America.
The stemless violets are a large group, with 20 species represented in NY. The sweet violet (Viola odorata) is the only non-native that has naturalized, though it is very common, especially in lawns, and is one of the first species to bloom in the spring. The only species in this group that has yellow flowers is the round-leaved violet (Viola rotundifolia), making it easy to identify in flower, though the round leaves are often overlooked later in the growing season. A few species have white flowers and these can be more difficult to identify, and the most challenging to separate are usually V. pallens and V. blanda. Occasionally blue violets can have white flowers as well. But the focus here is going to be those that have blue flowers.
So how many of the violet species that occur in NY are stemless blue violets? Based on what is recognized in the atlas, 14 of the 29 total species would fit into this group, which is just a little more than half of all Viola spp. This makes this group a challenge, as many are similar in appearance to the common blue violet (Viola sororia). Identifying the V. sororia look-alikes are beyond the scope of a single blog post, so I am going to focus on those that have deeply divided leaves. This is a small group that is frequently misidentified because there is a tendency to just call them all V. palmata because the leaves are palmately divided.
This brings the list down to 4-5 species:
- Viola pedata
- Viola sagittata
- Viola palmata
- Viola brittoniana
- Viola subsinuata
The one with the most unique flowers is V. pedata (bird’s foot violet), which has a limited distribution in NY, found at least historically on Long Island, Staten Island, and in Queens and New York Counties. There are also sparse records from Erie and Albany Counties that should probably be rechecked. The preference is for soil that is on the acid side and with excellent drainage, so sandy or rocky upland soils are most typical for this species.
In flower it is easy to identify as this is the only species of stemless blue violet with palmately divided leaves that has petals that are completely hairless. The petals also tend to be a softer blue than the other species, with the spur petal often with some white near the base and purple veins. Occasional plants will have the upper two petals a dark blue-purple, a striking form (f. bicolor) that I have not seen personally. The orange stamens are also conspicuous in this species, being more exserted from the flower.
Viola sagittata sometimes keys out close to the other species treated here, though any lobes are restricted to the base of the leaf blade and these are often small enough to be interpreted as teeth rather than lobes. This species has two varieties, var. sagittata and var. ovata, the latter of which is sometimes treated as a distinct species (V. fimbriatula). Early season leaf blades are said to be similar in both varieties and leaf shape is probably not a reliable characteristic when the plants are producing chasmogamous flowers. Mid-season leaf blades of var. sagittata tend to be less densely pubescent, often glabrous, and are more strongly sagittate, hastate, or cordate. Petiole length has been used to differentiate the varieties (those of var. ovata said to be shorter), however petiole length has been found experimentally to vary with available light, with plants receiving more sunlight developing considerably shorter petioles relative to blade length than those growing in woods.
Possibly the most reliable characteristic would be the sepals of the flowers, which are said to be ciliate in var. ovata and without cilia in var. sagittata. While this is best observed with a hand lens, if consistent would seem to be a reliable characteristic to go by. Even though this species barely made the list, the two varieties are probably worthy of further scrutiny in New York.
Although there are only three other palmately-lobed species that are recognized in New York, the taxonomy gets really messy from here on out because the variability in form of the leaves has given rise to innumerable names of dubious validity. Without getting into all of them, worth mentioning will be some of the names that have been applied to what is now recognized as Viola palmata.
Viola palmata can be a little difficult to identify in mid-summer when the leaves can look all alike (look for smaller leaves near the base, as these are the spring leaves), as what sets this species apart from the others is that the leaves are heterophyllous (hetero = different, phyll = leaf). The first leaves to emerge in the spring are unlobed and resemble those of V. sororia, however by the time of flowering the plants will begin producing lobed leaves. Therefore, during or a little past flowering, plants should have a mix of lobed and unlobed leaves present. Unlobed leaves are said to be produced again late in the growing season.
Some Viola palmata plants can have leaves with many, deeply cut lobes and these have been called V. falcata or V. triloba var. dilatata. Such plants are now considered to just be a form (forma dilatata) of V. palmata. Plants with three primary lobes have been called V. triloba. The problem is that the depth and number of the lobes is so highly variable there is no definite demarcation. Therefore, all of these names (and many more!) have been combined under a single, variable species by most authorities. A search on JSTOR Global Plants for the name Viola triloba and V. palmata will turn up photos of numerous type specimens for other names that have been applied, all seemingly variations on the same general leaf form.
Two varieties of V. palmata are sometimes recognized, based mostly on whether or not the leaves and petioles are hairless (var. palmata) or pubescent (var. heterophylla). The former is also said to have the middle leaf lobe acute (pointed) and in the latter the middle lobe is blunt or obtuse. These varieties are not recognized in New York or New England.
The remaining species have, for the most part, homophyllous leaves. That is, the lobes of the early and later leaves are quite similar in pattern and appearance though the sinuses can be shallower, with lobes more like rounded teeth on the earlier leaves.
Viola brittoniana is primarily a coastal species that is limited to extreme southern New York. A single population is currently known in the state, on Long Island where it was once widespread. It has leaf blades and petioles that are essentially hairless, and slender sepals that gradually taper to the apex. The sepals also lack hairs on the margins (eciliate). This species prefers acid, sandy soils.
A close relative of V. brittoniana but with toothed rather than lobed leaves is V. pectinata, and where the two species occur together they are said to hybridize. It has only been collected in Nassau and Richmond Counties in New York and the most recent records are from the early 1900’s, so it is now considered to be historic. The leaves of V. pectinata are similar to those of V. sagittata except they are heart-shaped rather than elongate and the teeth extend further up the blade.
Finally, we have Viola subsinuata, which many people confuse with the heterophyllous species V. palmata. This species is similar to V. brittoniana but has sepals that are usually ciliate (hairs along margin) and that are relatively wide and round or obtuse at the apex. It also tends to be hairier, with pubescent leaf blades (at least on the underside) and petioles. It can be found scattered in the state on richer soil, often associated with limestone. It is similar to many forms of V. palmata but does not produce unlobed leaves in the spring (the lobes are shallower, sometimes resembling rounded teeth, but are present in the first leaves).
That covers all of the lobed, stemless blue violets in New York, at least how they are currently recognized. Harvey Ballard (Ohio University) has proposed a number of new names (or resurrection of older names), two of which that might apply in New York according to an article he wrote for the Spring 2016 NYFA newsletter. One is V. baxteri, which is considered part of the heterophyllous V. subsinuata species complex and may replace that species in some parts of western New York. Photos of V. baxteri on a website maintained by Ballard seem to show leaves that are more deeply divided, similar to V. brittoniana, but with hirsute foliage and broad obtuse sepals. A search of iNaturalist turns up some observations of V. subsinuata (including at least one identified as V. palmata) in west-central NY, on the edge of the indicated range of V. baxteri, but these are not obviously different in appearance from typical V. subsinuata. Perhaps the best way to tell them apart is by the color of the cleistogamous seeds, which are said to be much lighter in color (ivory) in V. baxteri than in V. subsinuata (light brown).
The other he called Viola pseudo-brittoniana, and it was said to possibly occur in dry woodlands near the coast in southern NY. This name does not show up on his new website (https://people.ohio.edu/ballardh/vgpena/), so it is not clear if that name has changed or has been dropped. The absence of this taxon from his map for the subsinuata group suggests the latter. That’s about as far as I could get with that, so the currently recognized species in the atlas are probably the most sensible to go by for now.