1922] SEARS—TARAXACUM 429 
thorough studies of entire plants excavated from various habitats. 
After some fifty plants had thus been secured and studied, several 
hundred additional spud collections were made for confirmatory 
purposes. Subsequently all conclusions were tested rigorously 
by field observations extending over a period of fifteen months. 
By comparison of root development, color and texture of cortical 
rind, number and nature of withered petiole zones, and general 
condition of rosette, it was found that the age of a given plant could 
_ be told with considerable accuracy. With this technique as a 
basis, attention was centered upon the development of life history 
details and leaf form changes. 
LIFE HISTORY.—This procedure revealed an intimate connection 
between leaf variability and life history. ‘The first consideration 
is that strict accuracy forbids speaking of Taraxacum as a perennial 
plant without qualification. As stated by Stork, Taraxacum ordi- 
narily does not bloom the first year. Blossoming occurs freely in 
both species throughout the second year. At the end of the period 
of most active flowering (autumn of second year or spring of third) 
the vigorous production of flower buds forces apart the inner leaves 
of the original rosette with their axillary growing points. From 
such growing points arise numerous secondary rosettes. As a 
rule several of these persist, functioning essentially as new indi- 
viduals, in spite of their common root connection. Throughout 
the length of the root each secondary rosette has its own strand 
of vascular tissue. The individuality of the parent rosette can 
only be maintained by the cortical tissues, and these rather gener- 
ally become cleft by fission and cork ingrowth. Observation 
indicates that such cleavage is hastened by conditions favoring 
rapid growth. Cleavage tends to become complete throughout 
the root length. 
LEAF ForM.—The relations between this life history and leaf 
form is intimate and direct. The first leaves of a seedling (juvenile 
leaves) tend to be entire and smooth (fig. 1), later becoming more 
dissected and generally more hairy (fig. 2). In T. laevigatum this 
dissection in general reaches a higher degree than in 7. vulgare, 
although the latter species, under conditions favoring luxuriant 
growth and numerous leaves, finally may attain an extreme degree 
