136 THE SECOND-CHROMOSOME GROUP 



great difficulty of classification was due to a speck from which olive 

 had been lost by crossing-over. On one other occasion speck has been 

 found which seemed to be without olive. We believe that most of the 

 few flies classified as " speck not-olive" in the F 2 of olive by wild were 

 either young flies in which the olive was not yet developed sufficiently 

 to be surely classifiable or were fluctuants extreme enough to cause 

 trouble, neither of the two olives being sufficiently dark or constant in 

 color to be invariably separable from the wild type. 



Throughout our discussion of speck and olive it has been assumed 

 that the olive color which is nearly always seen in speck flies is due to 

 a separate gene, and while this is probable, the evidence is by no means 

 conclusive. It may be that the two characters are the products of a 

 single gene, and that the supposed cases of "speck not-olive" have 

 been due on occasion to wrong classification of the poor character 

 olive, or to the action of minus modifying genes, or to a new speck 

 allelomorph which differed in this regard from the old. A further 

 careful investigation would be required to settle this question. 



VALUATION OF OLIVE. 



Olive II is of no value for further \vork, since the character is not 

 sufficiently distinct from wild-type so that the normal fluctuations in 

 these two body-colors do not overlap, and classification is accordingly 

 both difficult and inaccurate. Olive is subject to a general objection 

 to the usefulness of all faint body-colors, namely, the fact that body- 

 colors take some tune after hatching for their full development, and in 

 cases where the final difference is not great the intermediate stages are 

 unpleasant to work with and a source of error. A further defect in 

 the value of olive is the uncertainty as to whether the character may 

 not be found to be only another expression of the speck gene. Further- 

 more, in case the investigation should prove that two linked but sep- 

 arable genes were involved, the usefulness of olive would not thereby be 

 improved, since olive could be used for no purpose that could not be 

 better met by the use of speck or one or another of the mutations whose 

 genes are in this same region. As far as we have been able to observe, 

 the presence of olive has had no detrimental effect upon the viability 

 or other qualities of speck. 



TRUNCATE (T). 



(Plate 6, figures 1 to 6.) 



ORIGIN OF TRUNCATE. 



One of the early mutations (beaded, May 1910) had been run for 

 seven generations in stock cultures when a fly appeared (August 1910) 

 whose wings were somewhat obliquely truncated and somewhat 



