526 
M. G. Sykes 
chiefly examined but somatic nuclei from all other parts of tbe 
flower were also found in all stages of division. Anaphases and 
telophases were tbe most favorable stages for counting (Figs. 8 — 11), 
but tbe later anapbases and tbe prophases were more troublesome 
since tbe larger chromosomes are tlien very long and thin, while 
tbe smaller ones are often connected together in twos and tbrees by 
tbeir ends 1 2 3 ), (Fig. 3). 
In most cases it seems probable that I have rather underesti- 
mated tbe number of tbe cbromosomes, but it was always possible 
to count at least tbirtv six. It was generally very difficult to decide 
on anv definite figure between tbirty-six and forty-eigbt, but an 
approximation to tbe latter was usually found. Fig. 3 represents 
tbree successive sections of tbe same nucleus, it does not seem pos- 
sible to suppose that less than tbirty wbole cbromosomes, (fifteen 
pairs) are distributed through tbe tbree sections, wbile at least twelve 
to sixteen cbromosomes can be constructed from tbe cut portions. 
In Figs. 8 — 11, from forty to forty-eigbt cbromosomes are represented 
in each nucleus. Tbe following table gives tbe results of a series of 
countings; in a few cases one or more cbromesomes were also present 
in another section of tbe same nucleus. 
Table of countings. 
46 
49 
45-48 
46 
45—? 
47-48 
47—48 
47 
371 sj 
35 1 ' 
49 
49 
40 3) 
44 
44 
51 
The paired arrangement of tbe cbromosomes is extremely stri- 
kiug, as was sbown in some earlier figures 4 ). Botb long and short 
cbromosomes undergo considerable contraction before division takes 
place and on tbe equatorial plate they are always more or less easily 
sorted into two sizes. 
*) Cf. M. G. Sykes; »Nuclear division in Funkia«, 1908. Archiv f. Zellf. I. 
S. 8-9. 
2 ) See Fig. 7 a and description. 
3 ) A cut nucleus. 
4 ; M. G. Sykes 1, c. p. 13 and Figs. 4, 7. PI. I. 
