502 
AMERICAN JOURNAL OF BOTANY 
[Vol. io, 
but the rate of increase in size during a given period of years is not much 
in excess of the values recorded for Pinus Strobus. Furthermore, the ray 
initials in certain dicotyledons may have a maximum enlargement in 
diameter of from 30 to 40 microns, but the variability in the size of their 
fusiform initials is much less than that which occurs in most conifers. 
It is evident, accordingly, that, although the increase in size of the 
cambial initials, during the earlier stages of the enlargement of certain 
plants, is by no means a negligible factor, the rapid increase in girth of the 
lateral meristem must in general be due largely to a progressive increase in 
the number , rather than in the size , of its constituent cells. 
What then is the relative significance of the increase in the number of 
the two types of cambial initials? In the young shoot of Pinus Strobus , 
the combined diameters of the 70 ray initials form an arc of 980 microns, 
whereas those of the 8,796 ray initials in the 6o-year-old stem constitute 
an arc of 149,532 microns, or approximately one eighth of the total circum¬ 
ference of the cambium. Many of the dicotyledons have a much higher 
percentage of ray initials. Indeed, in extreme cases more than one half of 
the circumference of the lateral meristem may be occupied by ray initials. 
Therefore, in discussing the modus operandi of the increase in girth of the 
cambium, it is essential to distinguish between the effects of (1) those anti¬ 
clinal divisions which are concerned in the origin and multiplication of ray 
initials, and (2) those which produce an increase in the number of fusiform 
initials. 
The reader should bear in mind in this connection that Nageli’s generali¬ 
zation is based upon the assumption that the divisions in both types of 
initials are radial, whereas the conclusions of Hartig, Klinken, and Neeff 
refer to the activity of the fusiform initials. The writer will likewise confine 
his attention in the following pages to the fusiform initials, reserving further 
discussion of the ray initials for a subsequent paper of this series. 
Significance of the Arrangement of Fusiform Initials 
In the gymnosperms and less specialized dicotyledons, the fusiform 
initials are not arranged in regular rows, whereas in certain of the more 
highly differentiated dicotyledons they are symmetrically grouped in 
parallel, horizontal series. The question suggests itself, accordingly, is this 
striking dissimilarity in the architecture of the lateral meristem indicative 
of fundamental differences in the growth and division of the fusiform 
initials, and, if so, what factors are concerned in the transitions from a 
non-stratified to a stratified arrangement? 
As shown in figures 4-6, Plate XXXVI, the “fusiform” initials in 
stratified meristems are roughly hexangular with long parallel sides and 
abruptly tapering ends, and the elements of adjacent horizontal series do 
not overlap to any considerable extent. In other words, the form and the 
arrangement of the initials indicate very clearly that the increase in girth 
