Tlie Behavior of tlie Chromosoines as Studied through Linkage. 251 



are also on lecord. I liavu jtublisbed (Sturtevant '13c) some Fs data 

 showing the independence of a few genes in chromosome II from P and 

 Eb in chromosome III. In order to get really good data on this [nnut, 

 as in all cases of linkage, it is desirable to have l)ack-crosses, and there 

 appear to have been only tlncc of these published — one Iiy Mokkan ('12e), 

 one by Morcax and l?j(ii)(4i.:s ('13) and one by Sturtevant ('13c). 

 The results of these crosses I shall discuss below, in connection with 

 some similar results here presented for the first time. 



Since there is no crossing over, or at most only very little, in the 

 male between factors in the same chromosome, it makes no difference 

 which gene we choose from a chromosome for tests of the independence 

 of the three groups in spermatogenesis. Unfortunately, there is at present 

 only one such test recorded. Morgan ('l^e) has rejiorted certain crosses 

 made by Lipf, in which males of the constitution Pp were crossed to 

 pink eyed (pp) females. Since these males were also heterozygous for 

 the sex differentiator this is in effect a back-cross, and tiie offspring are 

 available for direct calculation of the "crossing over" ') between the I st 

 and 3rd chromosomes. When calculated by the usual method we ob- 

 tain the result 



1673/3341 = .50-070/0. 



Thus the 1 st and 3 rd chromosomes were obviously segregated to 

 the sperm quite at random. No such crosses are available at present 

 for the other two combinations (I & II, II & III), but the published Fa 

 data show that these chromosomes must be very nearly, probably 

 completely, independent. 



It is not quite so simple a problem to test the matter in the female, 

 since here we have crossing over within a given chromosome. The 

 method I have adopted is to use each end of chromosome I and II, and 

 either of the two loci in chromosome III. All possible combinations of 

 these have been tested by back -crosses. The results of these experi- 

 ments are shown in table XL 



From these results we may safely conclude, 1 think, that the three 

 chromosomes are distributed to the gametes quite at random. 



Before we conclude that they are really indei)endent, however, 

 there is another point which must be investigated. Does crossing over 

 in one chromosome have any effect upon the other chromosomes in the 



') As pointed out above tlie term crossing over is, of course, strictly applicable 

 only within a given chromosome, but it is convenient to extend it to such cases as this. 



