July, 1923] 
SANDS — STRUCTURE OF CHROMOSOMES 
347 
in the aceto-carmine preparations. Here only two size differences could 
be clearly seen, and this represents the most that could be counted. 
Strasburger (1904) assumes a compound construction for his gamosomes 
as he does for the Ide , but not quite in the same sense as presented in this 
paper. He says: 
Das Chromatin zieht sich aus den Lininfaden zuriick und lasst sie als wenig tingier- 
bare, zarte, perlschnurartig gegliederte Faden zuriick. Es bildet Kdrnchen, die sich um 
einzelne Zentren sammeln. 
The chromomeres of figure 8, Plate XXIX, seem to be made up of four 
smaller bodies each, but this is of course an optical section. Careful focus¬ 
ing up and down shows others to lie beneath. I have seen these smaller 
bodies again subdivided, but this is not always a mathematically regular 
phenomenon, such as Miss Merriman (1904) has reported for Allium. 
I am convinced of the accuracy of her figures 35, 39, and 41, Plate XII, 
but do not feel that she is warranted, without further evidence, in drawing 
the conclusion that each subdivision of the chromosome proceeds by fours 
any more than Bonnevie (1908) would be justified in such a statement 
by reason of her figures 11, 16, and 22, Plate XIV. 
The number of these subdivisions is variable. A careful study shows 
that the chromatic thread, contrary to the impression first given as it is 
seen in figure 13, Plate XXIX, is not made up of chromomeres side by 
side. The arrangement is more complex than this. In any given field 
of vision and with any group of chromosomes, the breadth of the individual 
chromosome, as well as that of any length of thread exclusive of constric¬ 
tion points, is constant, which would not be the case if they were made 
up of rows of bodies side by side, that is, if their real shape was ribbon-like. 
Heidenhain (1907) figures chromosomes which appear ribbon-like in 
cross section (text fig. 1). I have not seen such forms either in the living 
material or in that fixed by any of the methods mentioned above. 
In figures 3 and 4, Plate XXIX, an enlarged drawing of the ends of 
two threads is seen. Here one sees that the chromomeres are apparently 
arranged irregularly just inside the periphery of the linin cylinder. The 
surface of the cylinder may be nodular, irregular, or smooth, according 
to whether or not the chromomeres extend beyond the periphery. 
The chromomeres shown in figure 10 project more than those in figure 
5, so that the degree of irregularity presented by the chromosomal outline 
is a function of the degree of projection of these bodies. In the unstained 
and living chromosome, the surface has a punctate appearance, not only 
in the telophases but in the early anaphases or even before, so that, while 
Lundegardh (19126) has described a nodular chromosome structure in the 
telophases of the vegetative divisions, I would extend this description to 
include the reduction divisions and, in all cases, the chromosomes of the 
metaphases and anaphases. Some evidence of this is seen in figure 15, 
Plate XXX. Note the last chromosome at the bottom of the figure. 
