AMERICAN JOURNAL OF BOTANY 
[Vol. 9. 
A plant of this type obtained originally at Pierson's and set out in an 
open stock bed at the Brooklyn Botanic Garden underwent an interesting 
change in that it returned entirely, in crown and runners, to apparently 
the original type of elegantissima-compacta. There might be reason for 
suggesting that the growth in an open bench had something to do with it 
except for the fact that in many thousands of plants grown under the same 
conditions at Craig's I did not see any similar break. The explanation is 
probably found in the possibility that the particular number which I ob- 
tained from Pierson did not really represent a stable sport or mutation but 
rather a fluctuation of a type which will be described below. 
From the Wanamaker type there has developed at Pierson's a reversion 
to a completely once pinnate unruffled form, corresponding to Dwarf 
Boston" in general appearance and size (PI. X, fig. 5), but not sufhciently 
cultivated to warrant any general opinion as to its characters. It corre- 
sponds to the reversion from viridissima described above since it represents 
a second step back toward bostoniensis. 
In addition to the above mentioned, Trevillian has also detected among 
the plants in his charge another elegantissima-compacta mutation which 
approaches Piersoni in characters, i.e., it is twice pinnate and tall. Two 
examples from Pierson's have been grown at the Botanic Garden (PI. VI, 
figs. 2, 3). By reference to the illustiations it will be seen that there is 
some little difference in the shape and size of the ultimate segments of these 
two types. Both, however, are generally like the Piersoni-like reversions 
produced from superbissima, and, it may be added, like other similar rever- 
sions from progressive sports of a higher order of leaf divisions. A series 
of leaves of six of these types, produced by mutation from five different 
progressive sports representing several different grades of leaf division, is 
illustrated in Plates VI and VII. Thus figure i, Plate VI, shows a reversion 
from Smithi, a four-pinnate form. Attention is called to the fact that the 
six types, although alike in division, vary notice ibly in outline and carriage 
of the pinnae, in texture, and in minute characters. 
When it is recognized that each of the twenty derivative mutations from 
Piersoni noted on the chart may give rise directly or indirectly to reversions 
of a Piersoni type, the possibilities of a confused tangle of forms, practically 
impossible to differentiate by description, will be realized. What explana- 
tion could the ordinary systematic examination of such a group of forms 
bring forth? Some systematists would explain the variation as a set of 
closely related intergrading forms, connecting the extiemes as parts of one 
single "species." With fertile sexually reproducing forms, the suspicion of 
hybridism would certainly attach. Especially would this apply in such 
cases in which the reversion was intermediate in form and other characters. 
Obviously, if we may consider these vegetative mutations as analogous to 
variations among wild forms which appear similar, we may see adequate 
reason for caution in making generalizations regarding complexes of wild 
forms. 
