Viola

The genus Viola is very complex, in large part due to the frequency of interspecific hybridization. This is especially common within the Boreali-americanae (i.e., the acaulescent, blue-flowered species). Early-generation hybrids are usually intermediate in the characters that discriminate the progenitor species. Further, they very often grow is close proximity to the parents and are largely sterile (capsules usually contain only a few well-developed seeds). This is in contrast to occasional individuals that may display an odd character state (e.g., a plant of V. cucullata that has ciliate sepals but is otherwise typical of V. cucullata). Such plants may be later-generation products of past hybridization events. They sometimes occur within a single-species population and are probably best referred to as the species they most resemble (this approach is considered appropriate given that the morphology is an indication of the underlying genetic composition). Though many features of violet plants used for identification are straightforward, some are not intuitive and/or are rarely noted. Most species of Viola have two types of flowers and fruits: spring-appearing chasmogamous flowers and early to mid-summer-appearing cleistogamous flowers. Style morphology allows these two types of fruits to be differentiated. The chasmogamous flowers produce a capsule with an elongate style [Fig. 934], whereas the cleistogamous ones produce capsules with a tightly coiled style [Fig. 929]. Acaulescent species of Viola can be described as homophyllous or heterophyllous. Homophyllous species produce similar types of leaves throughout the season (in regard to lobed or unlobed), whereas heterophyllous species produce dissimilar leaves— the first ones are unlobed and later ones are lobed. Note that the first set of leaves to be produced by many acaulescent species with unlobed blades are not typical (i.e., they 
may differ in outline from the later leaves) and are not always provided for in the identification key. Haines (2001b) provided a relatively thorough review of the morphological characters used to identify members of the genus Viola. Viola hirsutula ×V. sororia was reported 
from CT by Kartesz (1999) based on Brainerd (1924). This hybrid was not explicitly attributed 
to New England by Brainerd, but V. hirsutula ×V. papilionacea was. Kartesz synonymized 
 V. papilionacea with V. sororia in part (and therefore the origin of the report of V. hirsutula ×V. sororia in CT). However, the correct application of the epithet “ papilionacea” has not 
been satisfactorily determined. Therefore, this report remains ambiguous until further work is done. Viola ×‌modesta House ( V. lanceolata ×V. primulifolia) was reported for MA and ME 
 by various authors; however, the specimens require further study to confirm their identities. References: Brainerd (1921, 1924), Russell (1965), Haines (2001b).