252 BOTANICAL GAZETTE [OCTOBER 
scintillans appeared as mutants, showing that the mutation coefh- 
cients for these three forms do not essentially differ from one 
another. 
Stray mutations into cana have appeared in later years in 
different cultures, as, for instance, in 1913 in those of O. laevifolia 
and O. scintillans. Three mutations from O. pallescens will have 
to be recorded in the pedigrees relating to this form. It seems 
probable that cana mutants also have appeared in previous years, 
but have not been distinguished from other narrow-leaved types, 
of which there have always been quite a number in the larger cul- 
tures. Many other mutations also have escaped observation 
during a series of years until a single specimen developed into 
a Strikingly new type. 
I have cultivated O. cana mostly as annuals, but in some 
instances as biennials. In both cases the stature is the same as that 
of O. Lamarckiana, but in the annuals the stems are slender and the 
foliage rather loose, whereas the biennials have thick and strong 
stems with dense foliage. The leaves are narrow, with a shorter 
blade and a longer petiole, and of a very striking gray color. The 
flower buds are long and thin, contrasting sharply with those of 
O. Lamarckiana and even more so with those of O. pallescens and 
O. Lactuca. The spike is less dense than in the parent species and 
the fruits are more cylindrical and narrower, containing fewer seeds 
(fig. 2). In the flowering condition, as well as in the stage of young 
rosettes, the plants are now easily recognized, but at other periods 
of their development it is often difficult to identify and count 
them, some specimens showing their marks very clearly, but others 
resembling more or less their Lamarckiana-like sisters. 
The easiest marks are afforded by the flower buds. Measured 
the day before opening and with the tube and ovary, their size 
varies, as a rule, from 7580 mm., against 80-95 mm. in O. Lamarck- 
iana cultivated under the same conditions; means 77.5 against 
90mm. The breadth, measured at the base of the conical part 
above the tube, is only 7 mm. The 4 tips at the top of the bud 
are more or less bent on one side, and this curious mark is so striking 
that it is often the first which draws the attention to a stray 
mutant of the cana type (fig. 2). The 4 lobes of the stigma are 
