12 AMERICAN FERN JOURNAL 
will serve to indicate the probably course of its evolu- 
tion. Smithi, its immediate ancestor, is a four-pinnate 
form, which presumably came from a form like Whit- 
mani or gracillima, if not from one of these. The pro- 
gressive decrease in leaf-length shown in the illustra- 
tion is probably not to be considered as progressive 
dwarfing, but merely as correlated with the increase in 
division. 
- The forms represented on the other two plates illus- 
trate variations which can only be briefly indicated here. 
All of them are Pierson forms (bostoniensis of course 
excepted), i. e., they have originated in the establish- 
ment of F. R. Pierson, Tarrytown, N. Y., which has 
been most prolific of new and interesting varieties. 
The origin of bostoniensis and Piersoni (Plate 2, Figs. 
1 and 2) has already been indicated. It may be added 
that Piersoni created a furore in the horticultural world 
when first introduced but is very little grown now owing 
to its tendency to revert, or throw once-pinnate leaves 
together with its twice-pinnate ones. The other two, 
elegantissima (Fig. 3) and elegantissima-compacta (Fig. 
4), are both direct sports from Piersoni in which a greater 
division of the leaf exists. They thus correspond to 
Whitmani in division. Elegantissima-compacta also illu- 
strates dwarfing. 
Plate 3 shows another dwarf sport from Piersoni, one 
in which no increase in division has taken place, super- 
bissima, (Fig. 3). Figure 5 of the same plate shows 
muscosa, a sport of superbissima, showing an increase 
in division and possibly also further dwarfing. 
The other figures of Plate 3 illustrate reversionary 
varieties of two different forms. “Dwarf Boston” 
‘Cf. “Some modern varieties of the Boston fern at their source,” 
— EE nse Garden, 16: 194-197, plates 161 and 162. — 
pekiniiadudaos ege of reproducing these two plates is hereby gratefully 
‘a, 
