SELECTION WITHIN PURE LINES OF PESTALOZZIA 157 



extreme plus and minus variates in the range of continuous variation. 

 From one point of view the method might be looked upon as an inefficient 

 one, in that it surely failed to take advantage of the total range of variation 

 in each generation. 



However, when it is considered that not only is the culture from which 

 the spore came known to possess the quality for which selection is being 

 made, but the culture derived from each spore in the line of descent is 

 known to possess this quality before that spore is selected as a parent, it 

 is clear that each selection is much more significant than is usual in cases 

 where many individuals are selected in each generation. For example, 

 in the line selected for plus spore length, each spore chosen as a parent 

 was known to be descended from a series of ancestral spores, each of which 

 had the potentiality for producing a longer-spored progeny than nine 

 other spores of its own generation, chosen at random. Similar plans of 

 selection have been used in one or two previous investigations but not 

 extensively nor with so full a knowledge of the behavior of both the 

 ancestors and the progeny of the parent individuals as in the present work. 



So far as the writer knows no other investigator of the selection problem 

 has been able to establish the mean of the offspring of each selected indi- 

 vidual as was done in this experiment. Most of the organisms which have 

 been used in selection experiments are of such a nature as to prevent the 

 possibility of securing a sufficiently large number of offspring, all of one 

 age and all developed under the same environmental conditions, to estab- 

 lish such means. Obviously to secure as many as a hundred direct off- 

 spring from one parent DifHugia, (JENNINGS uses the term parent as 

 applying to the individual of a dividing pair which retains the old shell), 

 one would have to wait for just that many divisions of the parent, which 

 would demand considerable time. The progeny being formed at different 

 times, might be subjected to different environmental influences, so that the 

 range of variation might be greatly increased. It is therefore not possible 

 to know, in such forms as Difflugia, what the mean of the offspring would 

 be were they all produced under the same conditions. When too small a 

 progeny is secured, one does not know whether these approximate the 

 mean value for the generation or lie at one or the other extreme of the 

 range. Pestalozzia is especially favorable in that one can attain a full 

 knowledge of each generation, which is not the case for most forms. 

 Though such data are obtained only by the expenditure of a vast amount 

 of labor it is believed they are well worth while in a serious study of so 

 complicated a problem as that with which we are here concerned. 



GENETICS 7: Mr 1922 



