[Vor. 11 
116 ANNALS OF THE MISSOURI BOTANICAL GARDEN 
theless, from the foregoing discussion, a probable line of descent 
may be surmised. 
зэм: Mimulsstrum Pseudoence 
Paradanthus 
= р Eunanus 
Eumimülus : 
; ОЕпое 
Erythran the | 74 
| Diplacus 
Tropanthus 
AnGeStral type 
Fig.2. Phylogenetic chart of the sections in Mimulus. Тһе solid lines indicate 
probable direct descent. Тһе dotted lines show the relationship of the composite 
мз to the other sections. 
There is nothing in the genus to disprove the assumption that 
woody plants preceded herbaceous perennials -and annuals. 
Tropanthus and Diplacus, shrubby forms with regular or irregular 
calyces, divided placentae, and corollas that are irregular but 
not personate seem in many ways to be the most primitive of 
our known groups. These two sections are not closely connected 
and there is no indication that one preceded the other, nor 
necessarily that all of the sections eame from either of these 
groups. The wide geographical distribution of Eumimulus, the 
stability of its vegetative and floral characters, and the close 
relationship of its two groups of species indicate that it is 80 
ancient group; whereas the highly developed personate corolla 
and blue color suggest considerable advancement from the 
primitive condition. Тһе regular calyx with its equal teeth 
might postulate derivation from Tropanthus, yet the locali m 
distribution of the latter in a region far removed from 807 | 
member of Ештітшив does not warrant this supposition. 1655 | 
more likely that Tropanthus and Eumimulus are offshoots fro — | 
