PARTHENOGENETIC PURE LINE OF APHIS AVEN/E FAB. Ill 



2. The results of this work with Aphis avence Fab. are. we 

 believe, sufficient to warrant the following generalization: 

 Fluctuating variations in a parthenogenetic pure line of Aphis 

 avence, Fab., and presumably in all parthenogenetic pure lines, 

 are in general not dependent upon germinal variations, and for 

 this reason are not capable of increase or summation through 

 the action of continued selection. Or to put it in another way: 

 Fluctuating variability in a parthenogenetic pure line is devoid 

 of one of its most important causes when exhibited in higher 

 animals that reproduce sexually that is, germinal variability. 



3. Corollary.- Some investigators, not considering the original 

 limitations that were placed on the pure line theory, i. e., that 

 it applies only to forms that reproduce asexually or by self- 

 fertilization, have attempted, and disastrously, to apply its 

 principles to sexually reproducing animals. This has been un- 

 fortunate, and has added confusion to controversy, and has been 

 wholly unnecessary. Castle's work with hooded rats 1 has shown 

 conclusively that germinal variability is to be assigned as the 

 cause, of a part at least, of phenotypical variability in these 

 higher forms, thus exempting them from the scope of the pure 

 line theory. 



4. Selection from extreme variants within a pure line of 

 Aphis avence Fab. is without effect upon the somatic characters 

 of succeeding generations. 



5. Long-continued selection from the extreme variant in each 

 succeeding fraternity produces no more of a change in the mode of 

 the variable than selection from individuals but slightly different 

 from the mean of the line and for only a few generations. Neither 

 produces any perceptible change. 



6. The pure line theory applies to parthenogenetic arthropods 

 as well as to forms that reproduce by budding, fission, or self- 

 fertilization. 



7. Selections from extreme variants for 44 consecutive genera- 

 tions, using a character that is well known to be inherited in 

 higher animals that reproduce sexually, failed to shift in the 

 least the mean for the line. 



1 See Castle, W. E., 1915, "Some Experiments in Mass Selection," Amer. Nat., 

 Vol. XLIX., pp. 7U-726. 



