ASSOCIATION OF CHROMOSOMES IN DIPTERA 253 



and 110 represent such prophases, taken from a small bit of 

 somatic tissue, apparently ectodermal, in which practically 

 all of the nuclei contain a multiple group. These nuclei are 

 very easily identified by their size as well as by their chromo- 

 some number. No complete metaphase figures were found 

 among these particular cells, so I could not determine whether 

 the nuclei contained quadruple (48) or double (24) the normal 

 number of chromosomes. But the essential point is clear, that 

 in each prophase nucleus the chromosomes appear in only six 

 different aggregates, just as they do in ordinary prophases.^'* 

 This means that here each aggregate is composed of four or 

 eight chromosomes instead of the usual two. In the tetraploid 

 groups two of the four chromosomes are sister halves of the 

 other two, and hence are respectively similar to them in make- 

 up. But all four of these chromosomes associate in essentially 

 the same manner, i.e., paired chromosomes are indistinguishable 

 from sister chromosomes in their manner of association. It is 

 a natural conclusion, therefore, that the paired chromosomes 

 bear much the same qualitative relation to one another as do 

 sister chromosomes (that they are qualitatively similar) and 

 that their association is dependent upon, although not neces- 

 sarily caused by, this relation.^'' Such a conclusion is in har- 

 mony with the known facts of cytology and genetics which 

 indicate that corresponding maternal and paternal chromosomes 

 are similar in composition. 



If the paired association in diploid cells is an expression of 

 the same underlying forces which bring about the association 

 (synapsis) during maturation, the views here set forth are sup- 



1^ Only five aggregates are conspicuous because one of the six is composed 

 of the very small chromosomes. Strasburger ('07) obtained tetraploid groups 

 in chloralized root tips of Pisum, and Stomps ('11) found such groups occasionally 

 in Spinacia, but both authors describe the chromosomes as arranged in pairs 

 instead of tetrads. Evidence is lacking on the crucial (prophase) stages, how- 

 ever, and such metaphase figures as are given may readily be interpreted as in- 

 dicating association in tetrads, even though the association is not close. The 

 question should remain open until tetraploid prophases are studied in these 

 plants. 



"See Lundegardh '15, who has come to very similar conclusions respecting 

 the bivalent chromosomes of the heterotypic maturation division in plants. 



