1() ANNALS NEW YORK ACADEMY OF SCIENCES 



even though the shape of the chromosomes is normal and their num- 

 ber is normal or nearly so. PI. Ill, Fig. 31 shows a distribution into 

 five groups of 3, 4, 6, 6 and 7 chromosomes that are all quite normal as 

 to shape, the total being 26 or only 2 more than the normal for the four 

 microspores that normally develop. In PL II. Fig. 30 four groups com- 

 posed of 12, 16, 11 and IN chromosomes are to he seen, and nearly all the 

 chromosomes are of the normal rodlike form. Here the number of micro- 

 spores that are forming is normal, hut the numbers of chromosomes are 

 increased to a total of 63 instead of the normal total of 24 for all the 

 members of a quartet. In such cases there were either fragmentations 

 with divisions of chromatin units in both the first and the second meiosis 

 or the number of chromosomes that entered synapsis was greater than 12. 



IRREGULAEITIES IN THE SHAPE OF CHROMATIN" MASSES 



The abnormal shapes that chromatin masses or units very frequently 

 take and retain begin to develop in the late stages of the first division 

 ( PI. II, Fig. 18) and become well developed in the equatorial plates of the 

 second division (PL II, Fig. 25). In the latter stage irregular 

 and X-shaped chromatin masses are commonly seen. In the majority 

 of cases each of these appeal's to he composed of two daughter chromo- 

 somes that have remained attached at the middle and have failed to 

 complete a separation. Among the groups of such structures some may 

 be found with the two roils of the X separating quite as do daughter 

 chromosomes while others are much lohed and very irregular in shape. 

 To count these structures as double or as two daughters would in many 

 cases give a total of more than 60 instead of the normal number of 24 

 for all nuclei of a completed second division. 



When these X-shaped chromatin bodies are assembled into two groups 

 after the anaphase of the first division, they may appear as represented 

 in PL II. Fig. 2b. In this particular figure there are 1"! in one group and 

 13 in the other. At this stage the number tor the two groups frequently 

 varies around 30. It was a figure similar to this that led Helling ( 1 !»•.'•"> I 

 to believe that there are 33 chromosomes, or three sets of 11, in the 

 somatic cells of plants of the Europa Davlilv. Helling, however, made no 

 studies of the somatic divisions. 



It is perhaps a significant fact that the \-shaped and apparently 



double chromosomes have not been found when the total Dumber for the 



.,, ( ,iid division is oearly normal — that is when there are hut (i split- 

 ting chromosomes lor each of the two equatorial plates of the second 



