Viola (odorata or tricolor)

[ReneZ]

"This phytomorph (Fig. 4.142a) clearly shows linear, terminal stipular lobes, as in the North American native V. bicolor Pursh (V. rafinesquei Greene) (Fig. 4.142b)—not spatulate, as in the Eurasian V. tricolor L. Also, this phytomorph matches the blue flowers of V. bicolor, not the tricolored ones of V. tricolor; V. bicolor flowers are uniformly cream to blue, whereas those of V. tricolor usually have two purple upper petals and three cream to yellow lower petals ..."

There is no chance to decide from a medieval drawing which of two very similar species is meant. Simply from that aspect, both would be equally valid (or in fact invalid). Neither of the two can be preferred.
This is not my opinion. This is what one reads in all literature on late antique and medieval herbals. 

As you know, and hint at, there is another good reason why a North American plant is not expected in this MS.
That can best be kept outside this discussion. The related argument does not depend on the viola illustration.