676 
Journal of Agricultural Research 
Vol. XXVIII, No. 7 
It is 10 univalent chromosomes that seem to cause the irregularities observed 
in the fusing of the chromosomes, and in their distribution during the reduction 
divisions. The heterotypic (Pl. 1, H and I) and to a less degree the homotypic 
(Pl. 1, J) divisions are very characteristic. The bivalent chromosomes split 
in a normal manner and move toward the poles. This division seems also to 
include all or nearly all of the trivalent chromosomes, one part of each going 
toward one pole and two parts toward the other, although the double portion 
appears to separate at about the time the chromosomes move from the plate; 
consequently all move as single units to the poles. The univalent chromosomes 
are found scattered a bit apart from the nuclear plate while the bivalent and 
trivalent chromosomes are dividing. Following the separation of the bivalents 
and trivalents the univalents collect at the nuclear plate (Pl. 1, G). By the time 
the earlier-divided chromosomes arrive at the poles the univalents at the plate 
move, some toward the one, some toward the other pole (Pl. 1, H), without 
undergoing division. These undivided univalent chromosomes are frequently 
too late in their movement toward the poles to be included in the forming daughter 
nuclei and are left in the cytoplasm to degenerate (Pl. 1,1). A study of the 
homotypic divisions shows that occasionally chromosomes lag on the spindle, 
and some of them fail to be included in the daughter nuclei (Pl. 1, K). The 
outcome of such a meiosis is the very unequal distribution of the 30 chromosome 
units, seen at the heterotypic prophase, into the four resulting pollen grains. 
This type of meiosis resembles that described by Rosenberg (19), Kihara (12), 
and other authors for hybrids resulting from crosses between species differing in 
chromosome number. 
That the 10 univalent chromosomes, which lag on the spindle of the hetero¬ 
typic and homotypic anaphases, came from the corn parent is suggested by the 
pollen study of one of these bigeneric hybrids. The study was made after it had 
been observed that the hybrids between starchy and waxy strains of maize, as 
well as hybrids between Euchlaena mexicana and waxy maize, produced pollen 
grains filled with starch and erythrodextrin in a 1:1 ratio. (See Table II.) The 
pollen of E . perennis X Chinese waxy maize showed only 3 or 4 per cent of the 
grains filled with erythrodextrin. If the 10 corn chromosomes had gone to make up 
the bivalent chromosome, approximately a 1:1 ratio of pollen grains filled with 
starch and erythrodextrin would have been expected. But if the 10 corn chromo¬ 
somes are the univalents that lag on the spindle of the heterotypic and homo¬ 
typic divisions and some of them fail to be included in the daughter nuclei, the 
result would be that a small per cent of the pollen would carry chromosomes 
having the erythrodextrin factor introduced by the corn parent. 
Table II .—Pollen counts of forms segregating for the starchy and erythrodextrin 
characters in pollen formation 
Plant 
Fertile 
Sterile 
Per cent 
of red 
in fertile 
Per cent 
sterile 
Blue 
Red 
Clear 
Starchy X waxy maize.._.. 
2,069 
2,046 
A few. 
49.7 
Waxy maize X E. mexicana . 
421 
392 
164 
48.2 
16.8 
Waxy maize X E. perennis . 
296 
13 
96 
4.2 
23.7 
Starchy coix X waxy coix..... 
658 
664 
472 
50.2 
26.3 
S. P. I. 48866. 
277 
290 
No count. 
51.1 
Starchy coix X waxy coix... 
127 
402 
No count. 
76.0 
S. P. I. 48867. 
521 
1,310 
2.582 
1 71.5 
58.5 
