120 
cases being confined to the end of the panicle branches. When it is further 
considered that also in Agrostis canina L. and other species of Agrostis there 
are found forms either with or without awns, it is obvious, it would seem, 
that the presence or absence of awns cannot be relied upon as a species 
character, as is also emphasized by Hitchcock (11, pp. 11-12). 
Although, in the writer’s opinion, neither the shape of the panicle 
nor the presence or absence of overground, creeping stolons can alone be 
deemed of sufficient taxonomic value to be used as a specific characteristic, 
it may be argued that a certain type of panicle, associated with presence 
or absence of stolons, may together furnish taxonomically good species 
characters. Indeed, the so-called Agrostis maritima Lam., which is char- 
acterized by having a dense, narrow panicle and overground, creeping 
stolons, is by some North American authors regarded as a distinct species. 1 
If a dense, narrow panicle always were associated with the presence of 
long, creeping, overground stolons, one might be tempted to segregate 
such forms as A. maritima Lam. as distinct species, but that is by no means 
the case. There exist numerous forms with overground, creeping stolons 
which have panicles with spreading branches (Plate XVII). On the 
other hand, forms without stolons exist in which the panicle is contracted 
and spike-like (Plate XVI, figure 2). That is to say, narrow panicles are 
not necessarily associated with presence of overground, creeping stolons; 
neither are open panicles correlated with absence of such stolons. In 
other words, the shape of the panicle and the stoloniferous habit are char- 
acters which are quite independent of each other. They might be com- 
bined in various ways, with the result that a large number of what generally 
are termed intergrading forms occur. 
As, furthermore, all the forms of the A. stolonifera group are perfectly 
fertile, producing normally developed pollen and setting seed abundantly, 
they can no more be rated as distinct species than can for instance the 
numerous types of Bromus inermis, but must be regarded as mere varieties 
of a single, very polymorphous species. 
Nomenclature. Up to quite recently the name Agrostis alba L. has 
commonly been applied to this polymorphous species or to certain elements 
of it by North American botanists. In 1918, however, Piper (27, p. 4) 
pointed out that Linnaei “description and references apply partly to wood 
meadow grass ( Poa nemoralis L.)” and he, therefore, rejected the name 
A. alba. L. A similar stand was also taken by Hitchcock (12, pp. 128-129, 
footnote) two years later. 2 
Both Piper and Hitchcock, in discussing the application of the proper 
scientific name to the various commercial bent grasses, make references 
to A. stolonifera L. Piper maintains that the latter name applies to 
Creeping bent which he considers specifically distinct from Redtop. He 
says (27, p. 4, footnote) : 
“This name (A. stolonifera L.), the first binominal given to any grass of the group 
here discussed, belongs to a plant growing about Upsala, Sweden, there known as Kryp- 
hven, that is ‘Creeping bent.’ It is quite intermediate in characteristics between Redtop 
and Fiorin as determined from authentic Swedish specimens secured from Dr. Carl Lind- 
man, of Stockholm, Sweden, ” 
1 It should be noted that in the original description Lamarck (17, p. 61) does not say anything about the stolon- 
iferous habit. Neither does Rouv (31, p. 63). Other authors, however, e.g. Ascherson and Graebner (l,p.l75), 
Hitchcock (12, p. 129), and Holmberg (14, p. 142), emphasize the presence of long, creeping, overground stolons. 
1 Already in 1898, Murbeck (25, p. 3) questioned the propriety of using the name A. alba. L. 
